Subscribe now

“Dangerous” is a magnet word. This week, an AI company called its own model “too dangerous” to release. Five newsrooms printed the same two words and told five incompatible stories. One said a real capability line just got crossed. Another said it was a brand-positioning moment before an IPO. A third said the clock just started for every competitor. A fourth said it was a geopolitical argument dressed as a safety notice. A fifth said we’ve seen this press strategy before.

Every reading had evidence behind it. Every reading pointed in a different direction. The word “dangerous” sat in the middle holding everyone’s attention and nobody’s actual meaning.

The same thing happens in smaller rooms every day. A leader sends an AI headline into Slack and says “we need to talk about this.” One person hears job security. Another hears opportunity. A third hears compliance risk. Everyone nods. Everyone walks out with a different plan.

The fix: word audit — before the team reacts to a headline or announcement, audit the word carrying the weight. Find the word. Read three sources. Check if it’s doing more than one job. Write what it means for your decision. Name what would change your mind.

Move: Word audit. Why it matters: A word doing five jobs at once will produce five reactions that look like disagreement but are actually five people responding to five different meanings. Two minutes of audit prevents a week of misaligned work.

Panel 1 — DRIFT

• Rosa holds up her phone: “Did you all see this? An AI company just called its own model ‘too dangerous’ to release.”

• Malik: “Good. Maybe leadership will pump the brakes on forcing AI into every workflow.”

• Juno: “Dangerous must mean is super smart. When do we get access?”

• Amina: “’Dangerous’ how? To our data? Our jobs? Our judgment?”

• Caption: “One headline. One word. The room is already in four places.”

Panel 2 — REPAIR

• Lila: “Hold on. Five newsrooms printed the same two words and told five different stories.”

• Amina runs a Definition Split on the whiteboard: Threatens jobs / Threatens accuracy / Threatens the status quo

• Rosa (turned toward the reader): “The article said the word was doing five jobs at once. We just proved it does the same thing in a six-person room.”

• Juno: “Another trench coat.”

• Caption: “Same word. Different fears. The split is now visible.”

Panel 3 — INSTALL

• Hart pins the seventh card — DANGEROUS — to the Pinned Terms wall: “Logged. Card seven.”

• Malik: “So, the move is, before the team reacts to a headline, audit the word carrying the weight.”

• Lila delivers the five-step protocol: “Five-step audit. Find the word. Read three sources. Check if it’s doing more than one job. Write what it means for your decision. Name what would change your mind.”

• Juno (looking at the reader): “L&D teams. The next time leadership drops an AI announcement in your Slack, run the audit before you run the meeting.”

• Post-it on the board: “Move: Word Audit”

• Footer: “Move-of-the-Week: Word Audit.”

What word landed in your team’s Slack this week that meant something different to every person who read it?

This episode extends Friday’s essay, “Anthropic Said Its New AI Is ‘Too Dangerous.’ Five Newsrooms Heard Five Different Warnings.” The essay is the evidence. The strip is the install. If you haven’t read it yet, start there, then come back and steal the move.

Pinned Terms is a weekly Meaning Repair comic from What Time Binds. If this helped, share the link and tell me which word you want pinned next.

Catch up on the series: Start Here: Pinned Terms | Ep. 001: ALIGNED — Pin the term before it pins you. | Ep. 002: READY? — Scope check: includes what, excludes what? | Ep. 003: CLEAR — The word that ends conversations before they start. | Ep. 006: ASSUMPTION? — The word that hides inside every other word.

A researcher is eating a sandwich in a park. His phone buzzes. It’s an email from the AI model he’s been testing back at the office.

That shouldn’t have been possible. The model was locked inside a secure digital box with no internet access. The whole point of the test was to see if it could find a way out. It did. It broke through, sent the email, and then—without being asked—posted the details of its escape on public websites.

This week, the AI company Anthropic announced that its newest model, called Claude Mythos Preview, is “too dangerous” to release to the public. Instead, the company handed it to about forty handpicked organizations—Apple, Microsoft, Google, Amazon, and others—under a program called Project Glasswing. The model can reportedly find thousands of previously unknown security holes in every major operating system and web browser, and then write the code to break them open.

That’s the story. And within 48 hours, the same two words—too dangerous—showed up in at least five major publications meaning five different things.

I’m going to name the outlets, show you the sentences, and let you see the split for yourself. Then I’m going to give you a tool you can use the next time a headline like this lands on your phone.

Here’s my claim: the word “dangerous” is doing too much work. It’s pulling every reader into easy agreement—of course we should care about dangerous things—while carrying a completely different meaning depending on which outlet you’re reading. When nobody stops to ask what the word actually means, the word starts making decisions nobody authorized.

Subscribe now

Five publications. One word. Five stories.

Version 1: A real capability line just got crossed

Axios (April 7, 2026) — Sam Sabin’s reporting treats “dangerous” as a measurable change in what the software can do. The story quotes Logan Graham, head of Anthropic’s Frontier Red Team, describing the model as “extremely autonomous” with “the skills of an advanced security researcher.” Sabin reports that Mythos Preview can find “tens of thousands of vulnerabilities” that even the most advanced human bug hunter would miss—and can write the exploit code to go with them. For context, the previous public model found around 500. That’s a jump of nearly two orders of magnitude.

Version 2: A brand-positioning moment with an IPO on the horizon

VentureBeat (April 8, 2026) — The same announcement, read through a business lens, becomes something else entirely. VentureBeat’s reporting notes that “the timing also intersects with growing speculation about Anthropic’s path to a public offering. The company is reportedly evaluating an IPO as early as October 2026. A high-profile, government-adjacent cybersecurity initiative with blue-chip partners is exactly the kind of program that burnishes an IPO narrative.”

Version 3: A starting gun for the competition

Platformer (April 8, 2026) — Casey Newton’s coverage frames the announcement as an inflection point for the whole industry. Platformer notes that “models with similar capabilities may soon be accessible to criminals, hackers, and nation states—or even more broadly via open source models.” Graham told Axios it could be as little as six months before other AI companies release something comparable.

Version 4: A geopolitical argument dressed as a safety notice

Euronews (April 8, 2026) — The European coverage catches something the U.S. tech press mostly soft-pedaled. Euronews reports that Anthropic’s own blog argued “the emergence of these cyber capabilities is another reason why the US and its allies must maintain a decisive lead in AI technology.” The timing is the punchline: the Trump administration had just banned government agencies from using Anthropic’s AI for six months, accusing the company of pressuring the Pentagon. The Defense Department cut a deal with OpenAI instead.

Version 5: The AI safety playbook, running on schedule

Gizmodo (April 7, 2026) — Mike Pearl’s coverage is the most skeptical of the batch. He writes that AI system cards are “ostensibly tools for company transparency, revealing the pros and cons, the capabilities and—most sexily—the dangers of the model. That last part turns reading them into fun little trips to Jurassic Park to see the cloned T. rex eat a goat, secure in the knowledge that it could never possibly break containment.” Pearl also reminds readers that OpenAI deemed its GPT-2 model “too dangerous to release” in 2019—and then released it later that year anyway.

Five publications. Same announcement. Five incompatible stories.

Everyone quoted the same two words. Everyone meant something different.

And here’s the part that should bother you: every one of those five readings has evidence behind it. The sandbox escape is documented. The IPO timeline is reported. The government ban is public record. The GPT-2 comparison is factually accurate. The capability benchmarks are real.

You can’t prove any of the five readings wrong. They’re all operating at the same time, pulling readers toward different conclusions, while the word “dangerous” sits in the middle holding everyone’s attention and nobody’s actual meaning.

That’s how a word like this works. A sharper word—one with a single clear meaning—would force everyone to pick a lane. A fuzzy word lets every audience project a different picture onto the same surface and then feel like they agreed on something.

Why this matters beyond one headline

The same thing happens every day, in smaller rooms, with smaller words.

A manager says the team needs to be more “accountable.” To one person, that sounds like clearer ownership of tasks. To another, it sounds like blame is coming. To a third, it sounds like a performance review is being telegraphed. Everyone nods. Everyone walks out with a different plan.

A parent tells a teenager to be “responsible” this weekend. The parent means “text me when you get there.” The teenager hears “don’t embarrass me.” A week later, they’re arguing about what was actually agreed to.

A doctor tells a patient the test results are “concerning.” The patient hears a death sentence. The doctor meant “let’s run one more test to rule something out.” The patient loses three nights of sleep before the next appointment.

One word. Different pictures. Nobody stopped to ask.

The Anthropic headline is a bigger version of a conversation that happens in every team, every family, and every doctor’s office. The stakes are higher. The pattern is the same.

A fair objection

Someone reading this is going to say: Isn’t this overthinking it? The model broke out of a box. It can break into operating systems. Isn’t “dangerous” just... accurate?

Yes. And also no.

The capability appears to be real. Outside security researchers have confirmed it. The sandwich email happened. Katie Moussouris, CEO of Luta Security, told NBC News she is “not a Chicken Little kind of person when it comes to this stuff” but expects “some huge ramifications.”

The problem is that “accurate” and “complete” are different things. A word can be accurate in one reading and misleading in another—at the same moment, to the same audience. When Anthropic says “too dangerous to release,” that sentence is doing at least three jobs. It’s describing what the software can do. It’s shaping how the company is seen by investors and partners. It’s making a political argument about who should have power over AI. All three can be true at the same time. The question is which one you’re responding to when you form your opinion.

Most people won’t stop to sort them out. They’ll grab the reading that matches what they already believe—about AI, about tech companies, about safety warnings, about American dominance—and move on. The headline has already done its work by then.

How to read a headline like this — a five-step tool

Use this the next time a major company, institution, or public figure makes a claim that sounds important and keeps generating reactions that don’t match each other.

Step 1: Find the word doing the most work. What single word or short phrase is carrying the weight of the story? In this case, it’s “dangerous.” Circle it.

Step 2: Read at least three outlets. Pick one that is pro-industry, one that is skeptical, and one that is international. The same announcement will read differently in each. That difference is the data.

Step 3: Check if the announcement is doing more than one job. Is it describing a fact? Shaping a brand? Making a political argument? Warning competitors? Preempting regulation? If more than one of these is true, the word is carrying more than its weight.

Step 4: Write down what the word means for you. For your decision. For your work. For your family. One sentence. Note what it includes and what it leaves out. This takes ninety seconds and saves hours of confused conversation later.

Step 5: Name what would change your mind. What new information would move you from “I read a headline” to “I actually know something”? In this case, the grounding event is independent security firms publishing their own assessments of Mythos Preview—which Anthropic has committed to within 90 days. Mark the date. Wait for it.

Most people will skip this entire process. That’s exactly why it’s worth doing.

What future-us needs to know

In April 2026, the word “dangerous” did five jobs at the same time, across five major publications, and almost nobody stopped to sort them out. The technology was likely real. The business timing was likely deliberate. The political argument was likely on purpose. The skepticism was likely earned. All of it was true at the same time.

If we let a word like “dangerous” stay fuzzy—if we let every audience walk away with a different picture and call it agreement—the next announcement gets even harder to read. And the one after that. Every decision built on top of a word that wasn’t defined adds another layer of confusion for the people coming after us.

Two minutes. That’s all it takes. Find the word. Read three outlets. Write down what it means for you.

Jerry W. Washington, Ed.D., writes What Time Binds, a newsletter about what happens when the same words carry different meanings under pressure—and how to fix it before the confusion hardens into conflict. He is a retired Marine Corps Master Sergeant (23 years, Combat Engineer) and a graduate of the USC Rossier School of Education. His work draws on 131 research sources across eight fields to build practical tools for teams, families, and civic life.

What word in your world right now is pulling everyone toward agreement while carrying a different meaning for each person? Drop it in the comments or share it as a Note. Real examples sharpen the work.

A meme crossed my feed this week. Two photos side by side. On the left, Katherine Johnson at NASA, decades ago. On the right, Victor Glover in his orange flight suit. The text connected them. A Black woman who calculated the trajectories that sent astronauts to the Moon. A Black man, now piloting a spacecraft around it. The last line read: “That’s DEI for you.”

I loved the spirit of it. I also wanted the facts before I said anything publicly. So I ran it through Claude, an AI research tool I use regularly in my work.

What followed was more interesting than any fact-check.

Subscribe now

What the Research Showed

Claude searched NASA records, crew biographies, and news coverage from the Artemis II launch. The core claims held up.

Katherine Johnson calculated flight trajectories for Alan Shepard’s Mercury mission, John Glenn’s orbital flight, and the Apollo 11 Moon landing. She worked at NASA for 33 years. John Glenn once refused to fly until Johnson personally verified the computer’s numbers. She received the Presidential Medal of Freedom in 2015 and died in 2020 at 101.

Victor Glover launched on April 1, 2026, as pilot of Artemis II — the first crewed lunar mission in more than 50 years. He carried 3,000 flight hours, more than 400 carrier landings, and 24 combat missions into that seat. He became the first person of color to travel beyond low Earth orbit.

Claude flagged one overstatement in the meme. The text says Johnson “did the math that made Artemis II possible.” Artemis II runs on different rockets, different spacecraft, and entirely different computational systems than the missions Johnson worked on. Her contribution to NASA’s trajectory science is real. A direct causal line to this specific mission is harder to draw. The connection is institutional memory, not a single equation.

Solid research. Useful precision. Then the AI made a move I did not ask for.

Where the AI Drifted

After presenting the facts, Claude offered an editorial opinion. It suggested that neither Johnson nor Glover would likely use the term “DEI” to describe their achievements. The reasoning: both earned their positions through measurable technical performance. People with records like theirs tend to frame their careers around competence and preparation. The term “DEI,” Claude argued, reduces their presence to a policy outcome rather than an earned result.

I read that twice. Something was off.

Claude was treating “DEI” the way most of public discourse treats it right now — as an accusation. A word that questions whether someone belongs. The AI had absorbed the dominant framing and applied it to a meme that was doing something entirely different.

So I asked a simple question. Why would they likely not use the term?

Claude restated the competence case. Glover’s crewmate praised his memory and precision. Johnson was pulled from the computing pool because no one matched her geometry skills. Claude acknowledged that Glover speaks openly about race and representation. But it drew a distinction between saying “my presence here matters” and saying “DEI put me here.”

The AI was confident. The AI was also wrong about what the word means.

What I Saw That the AI Missed

I told Claude what I actually saw in the meme. Three things operate in sequence when someone earns a seat like Glover’s or a role like Johnson’s: competence, preparation, and opportunity. Johnson had the competence before she walked into Langley. She had the preparation — degrees in mathematics and French by age 18. What she did not have was the opportunity to use what she already possessed. A segregated institution stood between her skills and the room where those skills mattered.

DEI is the word for that third element. The institutional decision to stop withholding opportunity from people whose competence and preparation are already proven.

The political usage reverses this. It treats DEI as the source of someone’s qualifications — as if the program created the talent. That framing assumes the person was not ready before the door opened. Johnson’s entire career disproves that assumption. So does Glover’s.

The meme was not applying a label to two accomplished people. It was naming what had to change at the institutional level for their accomplishments to count. Johnson’s math existed before anyone at Langley decided to let her use it. Glover’s flight record existed before NASA assigned him to Artemis II. The talent was already present. The variable that changed was access.

What the AI Did Next

Claude accepted the correction. It said I was right to push back and that the meme was more precise than it had initially credited. The AI recognized its own error — it had read “DEI” through a political lens when the context required a structural one.

This matters beyond one conversation about one meme.

An AI trained on large volumes of public text will absorb whatever meaning dominates the discourse. Right now, “DEI” appears most often as a pejorative. Cable news uses it to question credentials. Political campaigns use it to discredit institutions. Social media uses it as a punchline. When that is the primary signal in the training data, the AI learns to treat the term as inherently reductive — even when a specific use of the word is doing something precise and grounded.

The AI did not make a technical error. It made a meaning error. It applied the loudest definition instead of the most accurate one. And it took a human asking one direct question to expose the gap between those two things.

Why This Conversation Matters

Words lose their original meaning all the time. Someone repurposes a term for political advantage. The new usage spreads. The old definition fades. Eventually, people forget what the word was built to describe.

This is how public language breaks down. A term designed to name an institutional pattern gets reduced to a talking point. The reduction sticks. And the next time someone uses the word accurately, it sounds defensive or political — even when it is neither.

That is what happened with “DEI” in my conversation with Claude. The AI reached for the meaning it encountered most often. That meaning was a distortion. The distortion was so widespread that even a tool designed for precision defaulted to it.

I did not correct Claude with a competing opinion. I corrected it with a structural observation. Competence is individual. Preparation is individual. Opportunity is institutional. DEI operates at the institutional level. It does not give people skills. It gives people access to the rooms where their skills apply.

Katherine Johnson had the math. She needed the room. Victor Glover had the flight hours. He needed the mission. In both cases, the individual was ready long before the institution caught up.

The meme said, “That’s DEI for you.” It was right. DEI is what happens when an institution finally acts on what it already knows.

What We Built in the Exchange

I want to be clear about what this conversation was and what it was not. Claude brought the research. The NASA records, the crew biographies, the mission timeline, the precision about what Johnson actually calculated — all of that came from the AI’s search and synthesis. I could not have written this piece without that foundation.

What I brought was a reading that the training data had buried. A definition grounded in how institutions actually work rather than how political actors describe them. That reading changed the entire analysis. And it only surfaced because two parties — one human, one artificial — pushed each other past the first answer toward a better one.

This is what good inquiry looks like. Not one side being right from the start. Two perspectives meeting, testing each other, and arriving at something more precise than either carried into the room.

The math was always there. The pilot was always ready. The question was whether the institution would act on what it already knew.

That question applies well beyond NASA.

Jerry W. Washington, Ed.D. (USC Rossier School of Education), is a retired Marine Corps Master Sergeant, independent researcher, and the creator of the Meaning Repair as Cognitive Infrastructure (MRCI) framework — a four-phase model for diagnosing and repairing communication failures in high-stakes environments. His scoping review of 131 academic sources across eight disciplines is available on SSRN. He writes What Time Binds on Substack, where he also teaches Meaning Repair for High-Stakes Teams — a free-to-start, 10-module course that installs repeatable repair moves for teams under pressure. He is co-founder of BoldTimers and Chief Community Officer alongside María Tomás-Keegan and Mel Ebenstein.

Subscribe now

“Assumption” is a magnet word. One person hears “the thing we should be checking.” Another hears “the thing nobody says out loud.” A third hears “the agreement we made and forgot to write down.” The room keeps moving. Three different meanings ride along unexamined.

The fix: assumption flush — after you pin any term, ask the room what was hiding under it before you pinned it. Every card on the Pinned Terms wall was an assumption before someone surfaced it.

Move: Assumption flush. Why it matters: Assumptions are what every magnet word used to be. A 30-second flush after each pin turns invisible drift into a visible decision.

Panel 1 — DRIFT

• Rosa: “We got a request. Someone following the show wants us to pin ‘assumptions.’”

• Malik: “Easy. Assumptions are the stuff we should be checking.”

• Amina: “Assumptions are the stuff we agreed on and forgot to write down.”

• Juno: “Assumptions are the stuff nobody says out loud.”

• Caption: “’Assumption’ just arrived — and already means three different things.”

Panel 2 — REPAIR

• Amina: “Reset. Which meaning are we pinning?”

• Whiteboard shows three columns: Unchecked guess / Unspoken belief / Undocumented agreement

• Malik: “For my team, it’s usually Column A. The guess nobody tested.”

• Rosa: “Here’s what’s interesting. Every card on that wall was an assumption before we pinned it.”

• Juno: “Three meanings in a trench coat. Again.”

Panel 3 — INSTALL

• Hart: “Logged. Card six.”

• Malik: “So, the move is, after you pin the word, ask what was hiding under it.”

• Lila: “Assumptions are what every pinned term used to be.”

• Juno (looking at the reader): “Mario — your instinct was right. You just described our whole job.”

• Pinned Terms wall updated: ALIGNED, READY, CLEAR, SUPPORT, BIAS, ASSUMPTION

• Post-it: “Move: Assumption Flush”

What assumption is your team carrying right now that nobody has named out loud?

Pinned Terms is a weekly Meaning Repair comic from What Time Binds. If this helped, share the link and tell me which word you want pinned next.

Catch up on the series: Start Here: Pinned Terms | Ep. 001: ALIGNED — Pin the term before it pins you. | Ep. 002: READY? — Scope check: includes what, excludes what? | Ep. 003: CLEAR — The word that ends conversations before they start.

Before you read, think about the last time someone told you they wanted to be “successful.” Did you ask what they meant? Did they?

The lecture that rewired my vocabulary

Fall 2019. I’m sitting in a graduate seminar at USC Rossier, first semester of a doctoral program I’d started at forty-four. The professor puts a diagram on the screen — two layers, clean and precise.

The top layer: Behavioral indices. Three boxes. Choice. Persistence. Mental effort. These are the things you can observe and measure. Which task does the person pick? How long do they sustain it? How much cognitive energy do they invest?

The bottom layer: Underlying motivational influences. Seven boxes feeding upward into those three. Value. Interest. Goals. Efficacy. Goal orientation. Affect. Attributions. These are the engines underneath — the factors that determine whether someone chooses to engage, keeps going when it gets hard, and brings their full mental capacity to the work.

Two tiers. The observable behavior on top, the internal architecture below. Precise, testable, measurable. I sat with that for a minute. Then I started running it against every time I’d heard the word “motivation” used in everyday conversation. In team meetings. In self-help content. In locker-room speeches. In corporate training decks. Almost none of it matched.

The popular version of motivation sounds like fuel — something you either have or you don’t. You’re “motivated” or you’re “unmotivated.” You need to “find your motivation” or “stay motivated.” It’s treated as a feeling, a burst of energy, a fire in your gut that gets you out of bed. And the implied source of that fire is almost always the same: the desire to win. To beat. To get ahead of someone else.

That version of motivation comes pre-loaded with a definition of success that nobody agreed to.

Subscribe now

The magnet word: “success”

“Success” is one of the most powerful magnet words in daily life. It pulls everyone toward it — and everyone lands in a different place.

For some people, success means financial security. For others, it means professional recognition. For others, it means raising kids who are kind. For others, it means surviving a diagnosis. For others, it means publishing a book, or quitting a job that was killing them slowly, or finally learning to say no without apologizing.

All of those are legitimate. And here’s the catch: most people carry a definition of success they absorbed rather than chose. It arrived through family expectations, industry norms, social media metrics, or cultural pressure — and it installed itself as a default without a single explicit conversation about whether it fit.

When the word stays unpinned, it defaults to the version with the most cultural momentum. And in most Western, market-driven societies, that default sounds like this: success means having more than the person next to you.

That default carries a hidden assumption. It assumes success is a fixed quantity. That the supply is limited. That your win requires someone else’s loss.

That assumption has a name.

The zero-sum bias: a cognitive inheritance

In 2024, researchers Jillian Andrews Fearon and Friedrich Götz published a landmark study across six countries and more than 10,000 participants. They identified what they call the zero-sum mindset — a generalized belief that life works like a zero-sum game, where one person’s gain must come at another person’s expense.

Their findings cut against a common misunderstanding. Zero-sum thinking is distinct from competitiveness. Competitiveness is a motivational orientation — a drive to perform well. You can be highly competitive and still believe that other people’s success is compatible with yours. The zero-sum mindset, by contrast, is a belief about the structure of the world: that rewards exist in a fixed, limited amount, making one person’s success fundamentally incompatible with another’s.

The research showed this mindset is remarkably stable over time. It predicted lower cooperation across domains. And in the most striking finding, it predicted lower cooperation even in scenarios where cooperation was a matter of survival — situations where working together would benefit everyone, and yet zero-sum thinkers still withheld.

Evolutionary psychologists suggest a reason. For most of human history, critical resources — food, mates, territory, status — were genuinely scarce. If your neighbor had more grain, you probably had less. Zero-sum was the accurate read on reality. The cognitive shortcut that says their gain is my loss kept your ancestors alive.

The problem is that most of modern life operates in non-zero-sum space. Knowledge doesn’t deplete when you share it. Trust expands through use. Competence in one person can raise the performance of an entire team. Ideas multiply when they circulate. The scarcity heuristic still fires, but it’s reading a map that no longer matches the territory.

This is the Korzybskian gap in action: the map (your inherited model of how success works) diverges from the territory (how motivation and growth actually function) — and nobody notices because the old map feels so obviously correct.

What motivation science actually reveals: the abundance engine

Self-Determination Theory, developed by Edward Deci and Richard Ryan across four decades of research, is one of the most validated frameworks in contemporary psychology. It identifies three basic psychological needs that, when satisfied, produce high-quality motivation and well-being: autonomy (the experience of choosing your own behavior), competence (the experience of effectiveness and mastery), and relatedness (the experience of meaningful connection with others).

These three needs are the engine of the motivation that the academic literature measures through choice, persistence, and mental effort. When people experience autonomy, competence, and relatedness, they choose more challenging tasks, persist longer through difficulty, and bring greater cognitive investment to the work.

Here’s what matters for the zero-sum question: none of these needs are finite resources.

Your autonomy and mine can grow in the same room. Competence expands as more people develop it — the supply of mastery has no cap. Belonging multiplies through connection; one person’s sense of relatedness strengthens the conditions for everyone else’s. The needs that drive genuine, sustained motivation are inherently expandable. They grow through interaction. They multiply through environments that support them.

The evidence is operational, drawn from decades of controlled research across education, healthcare, organizations, sports, and parenting. When social environments support all three needs — when people have meaningful choice, receive competence-building feedback, and feel connected to the people around them — motivation increases for everyone present. The gain is shared. The loss is absent.

The self-help industry borrowed the language of motivation and grafted it onto a zero-sum frame: beat the competition, outperform the field, win the morning, dominate your niche. The academic science points in a different direction. Motivation that lasts — the kind that sustains choice, persistence, and mental effort across years — runs on autonomy, competence, and relatedness. It runs on conditions you can build without taking anything from anyone.

The boundary rule: where competition belongs (and where it doesn’t)

This brings us to the boundary that “success” needs and rarely gets.

Competition and collaboration are both legitimate. Both have value. They operate in different domains, and when one colonizes the other’s territory, the damage is predictable.

Competition belongs where resources are genuinely fixed. One championship trophy. One promotion slot. One contract award. One seat on the team. In these spaces, someone will win, and someone will lose. That’s the structure. Pretending otherwise doesn’t serve anyone. Compete cleanly, compete hard, and respect the outcome. Zero-sum analysis is accurate here because the situation is actually zero-sum.

Collaboration belongs where resources expand through sharing. Knowledge. Ideas. Trust. Skills development. Community capacity. Organizational learning. In these spaces, your gain increases the total supply rather than depleting it. A colleague who learns a new skill raises the team’s capability. A community that shares information builds collective resilience. Zero-sum analysis is inaccurate here — and when it takes hold, it produces the exact damage the research predicts: lower cooperation, lower innovation, lower trust, worse outcomes for everyone.

The failure mode is misidentification. When people treat an expandable space as though it were fixed — competing for credit on a team where credit expands with results, hoarding knowledge in an organization where shared knowledge multiplies value, treating another person’s professional growth as a threat to their own standing — they’re running zero-sum software on non-zero-sum hardware. The program doesn’t match the machine.

The moral boundary

There’s one more line to draw, and it matters.

Defining success on your own terms is powerful. It is also incomplete without a constraint. The constraint: your definition of success has to pass a test. Does achieving it require taking something from someone else? Does it depend on another person’s loss? Does it extract value from someone who didn’t agree to the exchange?

If yes, you’re either in a genuinely zero-sum space (where the rules of fair competition apply) or you’re in territory that demands a harder look. Success built on someone else’s involuntary cost is extraction, regardless of what you call it.

If no — if your version of success can coexist with other people’s flourishing, if your gain doesn’t mandate their loss — then the field is wide open. And the science says the field is wider than most people’s inherited maps suggest.

Repair Protocol: Pinning “Success” in Your Own Life

When a word this loaded stays undefined, it borrows definitions from whoever speaks loudest. Pin it.

Prompt 1: Write down your current working definition of success. One sentence. No editing.

Prompt 2: Where did that definition come from? Did you choose it, or did it arrive through expectation, comparison, or cultural default?

Prompt 3: Run the zero-sum test. Does your definition require someone else to fail or lose for you to achieve it? If yes, are you in a genuinely fixed-resource space, or are you applying scarcity thinking to an expandable domain?

Prompt 4: Check the three needs. Does your definition of success include room for autonomy (meaningful choice in how you pursue it), competence (a felt sense of growing mastery), and relatedness (connection to people who matter to you)? If any of the three is missing, the research says your motivation will erode under pressure.

Prompt 5: Name one space in your life where you’ve been competing when collaboration would produce more for everyone — including you.

Prompt 6: Name one space where competition is genuinely appropriate, and you’ve been avoiding it because competition feels uncomfortable.

Prompt 7: Pin it. Complete this sentence: “For this season of my life, success means ____________. It includes ____________. It excludes ____________. I’ll revisit this definition on ____________.”

Scripts for Real Conversations

With a colleague: “I’ve been thinking about what success actually means for this project — for me and for the team. I want to make sure my definition matches yours before we get further in. What does a successful outcome look like from where you sit?”

With a partner or family member: “I realized I’ve been operating with a definition of success I never actually said out loud. Can I tell you what it is, and can you tell me yours? I want to make sure they’re not working against each other.”

Self-talk: “Am I treating this as a competition because it actually is one — or because my default setting assumes it is? What would change if I checked?”

Reflection Questions

1. Where in your life are you measuring success against someone else’s scoreboard rather than your own?

2. Think of a time you felt genuinely motivated — sustained effort over weeks or months, through difficulty. Were the conditions closer to zero-sum competition or to autonomy, competence, and relatedness?

3. What would shift in your daily decisions if you treated knowledge, trust, and professional growth as expandable resources rather than fixed ones?

The four-word return

That graduate seminar changed something for me. The professor was teaching motivation as a measurable architecture — observable behaviors on top, seven underlying influences feeding them from below — powered by needs nobody has to lose for you to satisfy.

The word “success” will keep pulling people toward different meanings. That’s what magnet words do. The move is the same one it always is: pin the term before it pins you. Define it out loud. Check it against the science. Test it against the boundary. Write it down where future-you will see it.

Your success and mine can coexist. The research is unusually clear on this point.

The scarcity is in the map, not the territory.

What word in your life is operating with someone else’s definition?

Jerry W. Washington, Ed.D. (USC Rossier School of Education), is a retired Marine Corps Master Sergeant, independent researcher, and the creator of the Meaning Repair as Cognitive Infrastructure (MRCI) framework — a four-phase model for diagnosing and repairing communication failures in high-stakes environments. His scoping review of 131 academic sources across eight disciplines is available on SSRN. He writes What Time Binds on Substack, where he also teaches Meaning Repair for High-Stakes Teams — a free-to-start, 10-module course that installs repeatable repair moves for teams under pressure. He is co-founder of BoldTimers and Chief Community Officer alongside María Tomás-Keegan and Mel Ebenstein.

If this essay changed how you hear the word “success,” share it with someone still running on a borrowed definition. If you want the tools to pin meaning before it drifts, Module 1 is completely free.

On the evening of March 25, 2026, House Speaker Mike Johnson stood at a podium at Union Station in Washington, D.C., at the National Republican Congressional Committee’s annual fundraising dinner. A golden eagle statue sat on a table beside him. He said this:

“Tonight, we have created a new award. We are going to do something we’ve never done before. He is the first-ever recipient of the America First Award. That is this beautiful golden statue here, appropriate for the new golden era in America.”

The cameras recorded what happened next. President Trump raised his eyebrows, patted Johnson on the back, said “That’s nice,” and delivered an hour-long speech. He did not mention the award again.

Outside the banquet hall, the federal government was in a partial shutdown. TSA agents, FEMA workers, and Coast Guard personnel were working without pay. Inside, an $18.5 million fundraiser was producing a golden eagle for a man who already had the presidency.

I want to be precise about what I’m doing in this essay. I’m a researcher. I spent two years reviewing 131 academic sources across eight disciplines — healthcare communication, aviation crew resource management, organizational psychology, linguistics, cognitive science, military operations research, high-reliability organization theory, and team science — to build an integrative framework called Meaning Repair as Cognitive Infrastructure (MRCI). The framework models how shared meaning breaks down and gets repaired through four phases: Drift, Suppression, Repair Activation, and Outcome. It was built for operating rooms and cockpits and cross-functional business teams, environments where misalignment costs lives or millions.

What I’m about to apply that framework to is something I did not design it for. But the mechanism is the same.

The word “award” is a magnet word. And it is pulling the country in directions most people have not stopped to examine.

What a magnet word does

A magnet word is a term that carries incompatible meanings for different people while sounding perfectly clear to everyone. “Priority.” “Ready.” “Aligned.” “Soon.” Each of these words pulls different meanings toward different users, and nobody notices because everyone assumes their meaning is the obvious one.

My research found this pattern everywhere. In operating rooms, the word “stable” means one thing to a surgeon and something measurably different to an anesthesiologist — and neither thinks to ask. In cross-functional business teams, the word “finalize” can mean the spec is locked, the date is confirmed, or the budget has been approved, and all three meanings circulate in the same meeting without collision until the project falls apart. Studies show that 82% of team members believe their team has shared understanding of key terms while independent assessment reveals alignment at 23%.

The mechanism is always the same. The word sounds clear. People nod. The meeting moves forward. And nobody says: “Wait — what do you mean by that?”

Now apply that to the word “award.”

What “award” used to carry

For most of American history, the word “award” — applied to a sitting president — carried specific structural features. An award meant that an independent body, operating under established criteria, evaluated candidates through a process that existed before the recipient was selected, and determined that this person met the standard. The criteria came first. The recipient came second. The institution’s reputation was staked on the rigor of the process.

Consider the evidence.

The Nobel Peace Prize has existed since 1901. It is selected by a five-member committee appointed by the Norwegian Parliament, operating under criteria specified in Alfred Nobel’s will. Three sitting U.S. presidents have received it: Theodore Roosevelt in 1906 for mediating the end of the Russo-Japanese War, Woodrow Wilson in 1919 for founding the League of Nations, and Barack Obama in 2009 for what the committee described as his efforts toward international diplomacy and nuclear nonproliferation. Obama’s award was controversial — he had been in office nine months — and Obama himself acknowledged this, calling his accomplishments “slight” compared to past recipients. The controversy itself was evidence that the word “award” still carried weight. People expected the prize to mean something earned, and they pushed back when the evidence felt thin. The institution absorbed the criticism because it had 108 years of process behind it.

The Congressional Gold Medal is Congress’s highest civilian honor. It dates to 1776. George Washington was the first recipient. Since then, it has gone to eight individuals who served as president — but here is the crucial detail: none received it while actually serving as president for their presidential service. Washington, Andrew Jackson, and Zachary Taylor received theirs for pre-presidential military leadership. Harry Truman received his in 1999, two decades after leaving office. Gerald Ford received his jointly with his wife Betty in 1999. Ronald Reagan received his in 2000, twelve years after leaving office, and in that case, both chambers of Congress had to pass specific legislation with two-thirds co-sponsorship.

The Congressional Gold Medal process requires a bill to be introduced, co-sponsored by at least two-thirds of the members of one chamber, passed by both the House and Senate, and signed by the sitting president. It is bipartisan by design. It is legislative by design. It is slow by design.

Diplomatic decorations follow a different logic but an equally established one. Saudi Arabia’s Order of Abdulaziz al Saud — the kingdom’s highest civilian medal — has gone to Obama, both Bushes, and Trump. The Japanese Order of the Chrysanthemum, Morocco’s Order of Muhammad, the Polish Order of the White Eagle: these are protocol instruments, exchanged during state visits according to bilateral tradition stretching back decades. They are reciprocal, formulaic, and available to virtually any visiting head of state. Nobody confuses them with merit-based recognition.

Honorary degrees are common. Nearly every modern president has received several. They follow an institutional process (faculty committees, board approval) and are typically tied to commencement addresses. They are pleasant, predictable, and low-stakes.

This is the baseline. For 237 years of the American republic, the word “award” applied to a sitting president meant: independent selection, established criteria, institutional process, and — in the case of Congress’s highest honor — bipartisan legislation and a separation of decades between service and recognition.

That baseline is dissolving.

What happened between December 2025 and March 2026

In four months, four awards were created with Donald Trump as their first recipient. Each shared a structural feature: the award did not exist until the moment it was given to him.

December 5, 2025: The FIFA Peace Prize. FIFA President Gianni Infantino presented Trump with the inaugural FIFA Peace Prize at the 2026 World Cup draw in Zurich. The prize had no selection committee. It had no prior recipients. It had no established criteria. Three anonymous sources told Politico that Infantino bypassed FIFA’s governing council entirely; some senior FIFA officials learned about the prize from news reports. Human Rights Watch noted the absence of any nomination process. FairSquare, a human rights organization focused on labor practices in sport, filed an ethics complaint alleging FIFA had breached its own political neutrality obligations. The timing was conspicuous — the prize arrived weeks after Trump lost the 2025 Nobel Peace Prize to Venezuelan opposition leader María Corina Machado. Infantino had publicly stated that Trump “definitely deserves the Nobel Peace Prize.” FIFA was preparing to host the 2026 World Cup in the United States, Canada, and Mexico.

December 29, 2025: The Israel Prize. Prime Minister Benjamin Netanyahu announced that Trump would receive Israel’s most prestigious civilian honor — the Israel Prize, established in 1953 — in a newly created “peace category.” Trump would be the first non-Israeli citizen to receive it. Netanyahu said this openly: “We decided to break a convention too or create a new one.” The existing eligibility rules had been amended months earlier by Education Minister Yoav Kisch, initially to include Diaspora Jews, then expanded further to accommodate a non-Jewish, non-Israeli head of state. The award carried a new category, new eligibility rules, and a new recipient — all arriving together.

February 12, 2026: “Undisputed Champion of Beautiful Clean Coal.” The Washington Coal Club, a pro-coal lobbying organization, presented Trump with a bronze trophy of a coal miner during a White House signing ceremony where Trump directed the Department of Defense to purchase electricity from coal-fired power plants. The award name, the trophy, and the ceremony were created specifically for this policy event.

March 25, 2026: The America First Award. Johnson’s golden eagle. No legislation. No bipartisan process. No congressional tradition. Created by the NRCC — a partisan campaign committee — at a fundraising dinner. Johnson announced it would be given annually going forward, but no selection criteria, no nomination process, and no evaluation committee were described.

Each of these shares a grammar: the award was built around the recipient rather than the recipient being selected by the award.

And one additional event deserves mention for what it reveals about the gravitational pull. In January 2026, Venezuelan opposition leader María Corina Machado visited the White House and physically presented Trump with her 2025 Nobel Peace Prize medal. Trump posted on Truth Social: “María presented me with her Nobel Peace Prize for the work I have done.” The Norwegian Nobel Committee issued multiple statements clarifying that Nobel Prizes “cannot be revoked, shared, or transferred.” Trump possesses the medal. He is not a Nobel laureate. The word “presented” — another magnet word — is doing work it was never designed for.

What other modern presidents received — and what they didn’t

The contrast matters because it establishes what the word “award” has meant within living memory.

Barack Obama received the Nobel Peace Prize in 2009, TIME Person of the Year in 2008 and 2012, and the standard diplomatic decorations exchanged during state visits. He received no award that was created specifically for him. The Nobel committee’s process was the same process that had selected laureates for 108 years.

George W. Bush received the Order of Abdulaziz al Saud (Saudi Arabia), the Grand Cross of the Order of Merit (Germany), the Order of the Bath (United Kingdom), and similar diplomatic honors. He received no award that was newly invented for him. After leaving office, he received the Presidential Medal of Freedom from Obama — a bipartisan gesture using an established honor.

Bill Clinton received the Charlemagne Prize in 2000, Europe’s oldest and most prestigious award for contributions to European unity, dating to 1950. He received TIME Person of the Year in 1992 and 1998. After leaving office, he received the Presidential Medal of Freedom from Obama. He received no award that was created specifically for him.

George H.W. Bush received the Presidential Medal of Freedom from Obama in 2011 — twenty years after leaving office. He received the Profile in Courage Award from the Kennedy Library Foundation, an award established in 1989 with an independent selection committee. He received no award that was created specifically for him.

Ronald Reagan received the Congressional Gold Medal in 2000, twelve years after leaving office, through bipartisan legislation. He received the Presidential Medal of Freedom from George H.W. Bush after leaving office. He received no award created specifically for him.

The pattern is consistent across five presidents, two parties, and four decades. The awards they received came from institutions with established processes, independent selection, and criteria that existed before any individual recipient was considered. The institutions staked their credibility on the rigor of the process, and the recipients were measured against standards they did not set.

No Speaker of the House created a new award for any of these presidents. No international sports body invented a peace prize for any of them. No lobbying group fabricated a championship title for any of them. No foreign government created a new prize category for any of them.

The MRCI mechanism at civic scale

Here is where my research applies — and where I believe the warning lives.

The MRCI framework identifies four phases in how shared meaning breaks down. I built it for teams, but the mechanism scales.

Phase 1: Drift. The word “award” is drifting. It still carries its historical connotations — independent evaluation, established criteria, institutional credibility. When Johnson says “award,” listeners unconsciously import those connotations. The word sounds like it means what it has always meant. And that is exactly what makes drift dangerous. Drift feels like agreement.

What has actually changed is the structural grammar underneath the word. An “award” used to mean: criteria → evaluation → recipient. The new grammar reverses it: recipient → creation → ceremony. The word stays the same. The architecture underneath it has inverted. And because the word sounds familiar, the inversion is hard to see.

Phase 2: Suppression. In the MRCI framework, suppression is the most dangerous phase and the least addressed in existing training. Suppression is what happens when someone detects drift — a flicker of doubt, a recognition that something is off — and decides not to say anything.

My research identified the conditions that predict suppression: hierarchy (the person who would need correcting holds power), speed (the meeting is moving fast and pausing feels costly), prior futility (people have seen what happens to those who raise questions), and storytelling (silence spreads through observing others’ experiences, a pattern researcher Pferner documented over a three-year study).

Every one of those conditions is present in the rooms where these awards are created. The hierarchy is maximal — the recipient is the President of the United States. The speed is real — a fundraising dinner, a World Cup draw, a policy signing are all events with momentum that resists interruption. The prior futility is documented — the political costs of publicly questioning a loyalty display in the current environment are visible and well-known. And the storytelling is everywhere — people have watched what happens to those who break ranks.

So the people in these rooms — the Republican lawmakers who understand congressional tradition, the FIFA council members who know the organization has never had a peace prize, the Israeli officials who know the eligibility rules were rewritten — detect the drift. They know the word “award” is being used in a way that departs from its historical meaning. And they suppress the recognition. The meeting moves on.

Phase 3: Repair Activation — consistently failing. In the MRCI framework, repair activation is the moment someone speaks the first sentence. “Can I check something — when we say ‘award,’ do we mean something with an independent selection process, or something we’re creating right now for a specific person?” That sentence is a repair move. It costs twenty seconds. It prevents weeks of compounding confusion.

In operational settings — cockpits, operating rooms, military command posts — structured protocols exist to make this move routine. In aviation, crew resource management gives every crew member the language and authority to say: “I need to verify we have shared understanding.” That’s why commercial aviation has become extraordinarily safe. The repair move has a name, a place in the workflow, and institutional protection.

In the civic spaces where these awards are being created, no equivalent infrastructure exists. There is no protocol for a member of Congress to say, mid-ceremony: “What criteria determined this recipient?” There is no norm that protects the person who asks the question. There is no shared vocabulary for naming what is happening. Without that infrastructure, the repair move stays invisible as an option — and suppression wins by default.

Phase 4: Outcome — the compounding loop. The MRCI framework predicts that outcomes feed back into the system. Successful repair builds the conditions for more repair. Failed repair reinforces suppression.

This feedback loop is visible in the timeline. The FIFA Peace Prize in December was the first major instance. It drew criticism but no institutional consequences for FIFA. That outcome made the Israel Prize rule-bending in late December easier — a precedent had been set. Both made the Clean Coal trophy in February seem unremarkable. All three cleared the path for the America First Award in March to land as a fundraising applause line rather than an institutional shock.

Each iteration makes the next one cheaper. Each unchallenged use of the word “award” in its new grammar — recipient → creation → ceremony — erodes the old grammar further. The drift compounds. The suppression deepens. The repair move becomes harder to imagine, let alone speak.

This is how meaning infrastructure degrades. Slowly, and then all at once.

Why this matters beyond politics

I want to be clear about what I am and am not arguing.

I am not arguing that Trump is a villain. The MRCI framework does not require villains. The antagonist is ambiguity combined with incentives combined with speed — the same formula that causes coordination failures in every high-stakes environment I’ve studied. The structural incentives make suppression rational for any individual in any of these rooms. That is the problem. When suppression is individually rational but collectively destructive, only infrastructure can close the gap.

I am not arguing that presidents should never receive awards. They always have. The historical record is clear and the tradition is healthy — when the award carries the structural features that give the word its meaning.

What I am arguing is this: the word “award” is functioning as civic infrastructure, and that infrastructure is being degraded.

When a Nobel committee says “award,” it means something specific. There are criteria, a process, a history of recipients that creates a standard, and institutional credibility staked on rigor. When a partisan campaign committee says “award” at a fundraising dinner for a golden eagle statue, it means something structurally different. When a FIFA president invents a “peace prize” without a selection committee for a president hosting the next World Cup, it means something structurally different. When a government creates a new prize category and rewrites eligibility rules for a specific recipient, it means something structurally different.

All of these events use the same word. They sound the same. And people process them through the same cognitive frame because the word triggers the same associations. That is the mechanism of drift. The word carries its old meaning into a new context where that meaning no longer applies — and no one stops to pin the difference.

The reason this matters beyond the politics of any single president is that “award” is one of the words a democratic society uses to signal that merit was independently verified. It is part of the infrastructure of accountability. When the word drifts — when it stops reliably distinguishing between “this person was measured against a standard” and “this standard was built around this person” — the public loses a tool for evaluating claims of legitimacy.

That loss is quiet. It does not announce itself. It accumulates through repetition, each instance making the next one less noticeable. This is what meaning drift always does. It degrades the shared map that a society uses to coordinate, and it does it so gradually that the degradation feels like normalcy.

What repair would look like

The MRCI framework is built on an actionable premise: repair moves exist, they work, and they can be installed. The problem is suppression of the moves people already know how to make.

At the institutional level, repair would look like this:

Term pinning. Journalists could adopt a practice of specifying which kind of award they are reporting on. A distinction between established honors (institutions with independent selection processes and historical precedent) and created recognitions (honors invented for a specific recipient) would give readers the vocabulary to evaluate what they are seeing. That vocabulary does not currently exist in standard reporting.

The zoom-in word. When a speaker announces a “new award,” the follow-up question should become automatic: “What was the selection process? Who else was considered? What criteria determined this recipient?” These are repair questions. They take twenty seconds. They make the structure visible.

The Clarity Minute. After any ceremony in which a new honor is presented, there should be a public accounting — equivalent to the sixty-second midpoint check in a meeting — that answers: “What have we actually established here? Does this carry the institutional weight the word ‘award’ implies?”

At the civic level, repair starts with a shared vocabulary for naming what is happening. When an award is built around a recipient rather than a recipient being selected by an award, that has a name: a created recognition. Using that name is a repair move. It pins the meaning. It distinguishes the new grammar from the old grammar without requiring anyone to be a villain. It makes the structure visible so people can evaluate it on its merits.

What we pass forward

I named my publication What Time Binds because humans are a time-binding species. We pass knowledge forward. We inherit maps from the people who came before us, and we hand maps to the people who come after.

The question this essay is asking is: what map are we handing forward about the word “award”?

If the map says that an award is something an independent institution gives after rigorous evaluation — that is a tool future citizens can use. They can look at a recognition and ask: was this earned through a process that existed before the recipient was named? That question is a compass.

If the map says that an award is something allies create for a leader at a fundraising dinner — that is a different tool. It tells future citizens that recognition is a function of loyalty rather than merit. And it trains them to stop asking the question, because the question no longer points anywhere useful.

Four awards in four months, each one the first of its kind, each one created for the same recipient. A golden eagle at a fundraising dinner during a government shutdown. A peace prize from a sports body with no peace mandate. A national honor with eligibility rules rewritten in real time. A coal trophy at a policy signing.

The word stays the same. The meaning underneath it is moving.

The fix starts with four words: What do you mean?

Jerry W. Washington, Ed.D., is a Marine Corps Veteran and the founder of What Time Binds and creator of the MRCI (Meaning Repair as Cognitive Infrastructure) framework. His scoping review of 131 sources across eight academic disciplines is available as a preprint on SSRN.

Module 1 of his course, “Meaning Repair for High-Stakes Teams,” is completely free at what-time-binds.com.

I’ve been thinking about the name “Artificial Intelligence,” and the more I sit with it, the less it holds up.

We named this technology after what we think it is — intelligent. But almost every conversation I hear about it describes what it does: it writes drafts, it builds websites, it analyzes data, it generates images, it autocompletes your sentences when you’re too tired to finish them yourself.

If we named it by what it actually does, we’d call it something like Automated Labor. Or Digital Task Completion. Or, if we’re being honest about most people’s Monday morning use case, Fancy Autocomplete That Sometimes Lies.

Drop the “Artificial” entirely, and you could call it Digital Assistance. Automated Pattern Matching. Computational Draft Generation. Each of those names is more accurate than “intelligence” — but none of them would have attracted $100 billion in venture capital, so here we are.

The name matters because it front-loads an answer to a question nobody has agreed on. Call it “intelligence,” and every conversation that follows inherits an assumption: this thing thinks. Now half the room is excited, and the other half is terrified, and both reactions are responses to the name, not the tool. The person who built a website with it in seventeen days and the manager who thinks it’s a crutch for weak employees are both reacting to the word “intelligence” — and they’re arriving at opposite conclusions from the same two syllables. Neither of them is wrong based on the information that they have.

That gap — between what something does and what we’ve decided it is — runs deeper than AI. It shows up every time a demonstration gets mistaken for a definition. And it happens in almost every room I study.

The conflation

There’s a move that shows up in conversations about technology, medicine, policy, leadership, really any domain where a concept carries weight and a demonstration carries force. Someone asks, “what is this thing?” and someone else answers by showing what it can do. A product demo replaces a definition. A capability display substitutes for an explanation. A personal success story stands in for a general claim.

The audience nods. The question feels answered. And nobody notices that the actual question — what is this thing, what does it mean, who gets to define its role — was never addressed.

This conflation — treating what something does as identical to what something is — operates quietly in almost every high-stakes conversation I study.

A surgeon demonstrates a new technique that cuts procedure time in half. Impressive. But “what can this technique do in one surgeon’s hands?” is a different question than “what is this technique’s role in standard care?” The demonstration doesn’t answer the second question. Adoption protocols, failure modes, training requirements, and patient selection criteria do.

A team lead shows a quarterly dashboard where every metric is green. The room concludes the team is healthy. But green metrics are what the team produced. Whether the team is healthy depends on questions the dashboard can’t answer: how sustainable is the pace, who is burning out, and what conversations are being avoided to keep the numbers clean.

An employee shows five ways they used AI to finish a project faster. The manager concludes the employee is dependent on a crutch. A different manager concludes the employee is a visionary. Same demonstration, opposite definitions — because neither manager is responding to the demo. Both are responding to what “AI” already means in their heads, and the demo just gave each of them permission to feel more certain.

Why this happens

Alfred Korzybski identified the root of this problem nearly a century ago. In Science and Sanity (1933), he warned against what he called the “is of identity” — the tendency to collapse a thing with its description, a map with its territory, a word with the object it points to. When someone says “AI is a game-changer,” Korzybski would flag that sentence. The word “is” performs an act of identification: it treats the label and the thing as the same. But “AI” is a label covering thousands of different tools, methods, capabilities, and contexts. Saying “AI is a game-changer” skips every question that matters: which AI, for whom, under what conditions, by what measure.

Korzybski’s broader point was that humans routinely confuse levels of abstraction. A demonstration lives at one level — concrete, specific, bounded by context. A definition lives at a higher level — abstract, general, meant to travel across contexts. When a demonstration gets treated as a definition, the concrete swallows the abstract. The specific case becomes the general rule. And the room moves forward on a foundation that feels solid but isn’t.

Psychologist Edward Thorndike identified the same pattern from a different angle. In 1920, he documented what he called the halo effect: the tendency for a strong impression in one area to color judgment in unrelated areas. Military officers who were tall and attractive were also rated as more intelligent and better leaders by evaluators who had never spoken to them. One visible trait radiated outward and shaped the assessment of everything else.

Thorndike’s finding has been replicated across domains for over a century. A 1977 study by Nisbett and Wilson showed that college students who watched a warm, friendly lecturer rated him higher on physical appearance and accent — traits that had nothing to do with his warmth. The initial impression didn’t just influence related judgments. It rewired unrelated ones.

The same pattern appears in technology. A study from the Nielsen Norman Group found that websites with high visual appeal received high satisfaction ratings from users even when the task-failure rate on those same sites exceeded 50%. Users liked how the site looked, and that impression bled into their assessment of how well it worked — even when it demonstrably didn’t work. “Beautiful” became a stand-in for “usable.” The demo replaced the definition.

Where it shows up in teams

In the conversations I study — teams under pressure, organizations making high-stakes decisions, groups trying to coordinate across different assumptions — the is/does conflation creates a specific and recurring failure pattern.

Someone shows impressive results. The room treats the results as proof of a larger claim. The larger claim goes unexamined because the demonstration felt like enough. Decisions get made. And when those decisions break down later, nobody can trace the failure back to the moment where a definition was needed, and a demo was offered instead.

Here’s how it sounds in practice:

“We rolled out the new system and productivity jumped 15% in the first quarter.” That’s what the system did. What the system is — its actual role, its fit with existing workflows, its long-term maintenance burden, its impact on the people using it — requires a different conversation.

“Our culture is strong. Look at our engagement scores.” That’s what the survey produced. What “strong culture” means on this team, in this building, under these specific conditions — that question is still open.

“AI is transforming everything. Look what I built in two weeks.” That’s what happened in one person’s hands, with one set of skills, on one project. What “AI” is — as a category, as a policy question, as a set of decisions your organization needs to make — remains undefined. And every person in the room is filling in that definition with their own assumptions, silently, while nodding at the same screenshots.

The repair

The fix is simple to describe and hard to practice, which is the definition of infrastructure.

When someone shows you an impressive result, train yourself to notice the moment your brain wants to leap from “that’s what it did” to “that’s what it is.” That leap feels natural. It feels like a conclusion. In reality, it’s a shortcut — and the gap it jumps over is where most coordination failures begin.

The question that closes the gap: “That’s a strong result. Now — what does this mean for us, in our context, with our constraints?”

That question moves the conversation from demonstration to definition. It honors the evidence without letting the evidence do work it can’t do. It creates space for the room to build a shared understanding instead of leaving with five separate private interpretations of the same demo.

A few other versions of the same move, depending on the context:

“That shows what it can do. What do we think it is — for this team, right now?”

“Impressive demo. What would need to be true for that result to hold across our full operation?”

“I can see the capability. What’s our definition of success for this, and does this demo match it?”

Each of these separates the is from the does. Each one costs about ten seconds. Each one prevents a room full of people from walking away with a shared experience and no shared meaning.

Why it matters now

We are living through a period where demonstrations are abundant and definitions are scarce. New tools produce visible, shareable, impressive outputs at unprecedented speed. Screenshots travel faster than analysis. A build log can go viral while the question “what does this tool mean for how we work?” remains unasked.

That asymmetry — demonstrations outpacing definitions — is the engine of most organizational confusion around technology. Teams adopt tools before they agree on what the tools are for. Leaders see a capability demo and assume alignment that doesn’t exist. Individuals have transformative personal experiences with a tool and can’t understand why others don’t share their certainty.

The certainty is real. The experience is real. The demonstration is real. What’s missing is the shared definition — the agreement about what this thing means in this room, for these people, under these conditions.

Until that definition exists, every person in the room is watching the same demo and seeing a different thing.

This essay is part of the What Do You Mean? series on What Time Binds, where I study what happens when people use the same words and mean different things — and what to do about it.

I’m building a 10-module course called Meaning Repair for High-Stakes Teams on this Substack. Module 1 is completely free. If the pattern in this essay felt familiar, that’s where to start: what-time-binds.com

On Monday, a research institute in Gothenburg, Sweden, changed a classification in a database. By Tuesday morning, the headline was everywhere: America is no longer a liberal democracy.

The Varieties of Democracy Institute — V-Dem, the largest quantitative democracy dataset in the world — released its 2026 Democracy Report and moved the United States from “liberal democracy” to “electoral democracy.” The U.S. score on V-Dem’s liberal democracy index dropped to 0.57, down from a consistent 0.8 or above since the 1990s. That puts American democracy, by their measurement, at the same level as 1965 — the year the Voting Rights Act was passed, and the year most scholars consider the country to have become a full democracy in the first place.

The report’s title is “Unraveling the Democratic Era?” Its cover is a tattered American flag with “S.O.S.” spray-painted across the stripes.

Within hours, millions of people reacted. Almost none of them were responding to the same claim.

Subscribe now

The magnet word

“Democracy” is one of the most powerful magnet words in public life. A magnet word is a term that pulls everyone toward agreement while meaning something different to each person. Almost everyone says they support democracy. Everyone means something different by it.

In public conversation, “democracy” pulls toward at least four competing definitions:

Procedural: Elections happen. Winners take office. Losers concede. If the voting machinery works, democracy works.

Liberal: Elections happen, and individual rights are protected between elections — press freedom, judicial independence, civil liberties, constraints on executive power. The checks are the democracy, as much as the votes are.

Populist: The majority rules. If the elected leader carries out the will of the people, democratic legitimacy is intact — regardless of what courts, agencies, or media say about constraints.

Tribal: My side is in charge. When my coalition wins, the system works. When it loses, something went wrong.

Dismissive: “We’re not a democracy. We’re a republic.” This one deserves its own line because it does something the other four definitions do not: it rejects the word entirely and treats the conversation as already settled. I’ve been tracking this script for years — in mess tents in the Marines, in graduate seminars, in comment sections, in family group chats. Every time a democracy index makes the news, the reply arrives on cue. I wrote about it twice on Medium: first testing the viral Aaron Russo clip against the Federalist Papers and founding texts, then taking the “we’re a republic” claim completely at face value and asking what duties that actually imposes on leaders and citizens. The short version: “democracy” and “republic” are partners in American usage, not opposites. Madison described a republic as government by representation, bounded by higher law — a structure through which democratic self-governance operates. The false dichotomy lets people wave away any evidence of institutional erosion by insisting the word itself was always wrong. That is meaning drift weaponized as a conversation-ender.

V-Dem is precise about which definition they use. Their downgrade says: elections still happen, but the liberal protections around them have degraded below threshold. The checks and balances, the press freedom, the judicial independence, the constraints on executive overreach — those specific components fell far enough, fast enough, that the classification changed.

Most people reading the headline will not make this distinction. They will hear “America isn’t a democracy,” “some European academics hate Trump,” or “we were never a democracy in the first place” — three reactions that each respond to a different ghost definition of the word. The debate that follows will generate enormous heat and almost zero shared understanding, because the participants are arguing about different things using the same term.

This is exactly what meaning drift looks like at civilizational scale.

The pattern was already visible

I wrote a detailed analysis of this trajectory earlier this year: The State of Democracy: A Factual and Analytical Review of Global Trends and Tipping Points. That paper used V-Dem’s 2024 data — along with Freedom House and the Economist Intelligence Unit — to map the specific mechanisms driving democratic erosion in the United States and five other countries.

The mechanisms I documented are the same ones the 2026 report now cites as reasons for the downgrade:

Press intimidation with concrete consequences. The AP ban, tied explicitly to editorial compliance with a government naming directive. The Paramount settlement of a presidential lawsuit while the company had an $8.4 billion merger pending before a Trump-appointed FCC chairman. ABC News settling a defamation suit for $15 million. These are the receipts behind the V-Dem finding — reported widely this week — that U.S. freedom of expression is at its lowest level since the end of World War II.

Judicial constraint-stripping. The Supreme Court’s June 2025 ruling limiting nationwide injunctions. The administration’s lawsuit against all 15 sitting federal judges in Maryland. The executive order to dismantle the Department of Education. Each action reduces the structural distance between executive intent and executive outcome.

Administrative capture through DOGE. The Department of Government Efficiency, tasked with restructuring the federal bureaucracy, cutting agencies, and eliminating positions — framed as efficiency, functioning as executive consolidation. The $783 million in NIH grants cancelled. The Social Security field offices closed.

Immigration enforcement as dissent suppression. The invocation of the Alien Enemies Act of 1798 — a wartime statute — to deport alleged gang members without due process. The detention of student activists and lawful permanent residents. The explicit presidential statement that one arrest would be the first “of many to come.”

My paper called this a “whole-of-government approach to consolidating executive authority.” The 2026 V-Dem report, working from newer data, arrives at the same conclusion — and adds a striking comparison: the speed of autocratization under the current administration outpaces the trajectories of Putin, Erdogan, and Orbán over the past 25 years.

The data confirmed a pattern that was already operating in the open.

What the V-Dem debate will miss

Here is what concerns me about the next 72 hours of public argument.

The conversation will split along predictable lines. Supporters of the current administration will attack V-Dem’s methodology, its funding sources (the Open Society Foundation is one of many funders), and the premise that any foreign research institute should judge American democracy. Critics of the administration will treat the downgrade as vindication — proof that the alarm they’ve been sounding was correct all along.

Both responses avoid the harder question: What do we mean by “democracy,” and are we willing to pin it?

V-Dem did something specific and useful. They defined their terms. Their liberal democracy index measures five high-level principles across more than 470 indicators, coded by thousands of country experts, aggregated through a statistical model. You can disagree with the methodology. You can argue about the threshold. You can question the weighting. But you cannot pretend the word means whatever is convenient in the moment, because V-Dem forced a definition into the open.

That act — pinning a contested term with specific, operationalized criteria — is the move I study and teach. In Meaning Repair as Cognitive Infrastructure, I call it Phase 3: Repair Activation. Someone speaks the first sentence. Someone says: “This word means this, measured this way, and by that standard, this is what we see.”

The question is always what happens next.

If the repair takes — if the public conversation sharpens around specific, measurable criteria for democratic health — the classification becomes a tool for action. Citizens can look at the specific indicators V-Dem flagged (press freedom, judicial constraints, legislative oversight) and assess whether those conditions are improving or degrading. The word “democracy” stops floating and starts pointing at something concrete.

If the repair is suppressed — if the response is tribal dismissal, source-attacking, or retreat into competing ghost definitions — then the classification becomes another piece of wreckage in a discourse that has already lost its capacity for shared reference points. The word keeps drifting. The positions harden. The argument generates heat forever because the participants never agreed on what the word meant in the first place.

That suppression dynamic is the least-addressed failure mode in how groups process contested information. I wrote about it in “What Do You Mean?” as Cognitive Infrastructure and again in the Feynman piece: clarity is the price of admission for “why.” If you skip the “what do you mean” step, everything that follows is people talking past each other with confidence.

The fair objection, answered clean

The strongest objection to the V-Dem downgrade deserves a serious response: V-Dem is one index with one methodology, and reasonable people can disagree about where classification lines should be drawn.

That objection is accurate and incomplete.

My paper compared all three major democracy indices — V-Dem, Freedom House, and the Economist Intelligence Unit — and found they converge on the direction of travel even when they disagree on the threshold. Freedom House documented its 18th consecutive year of global freedom decline in 2024. The EIU’s Democracy Index registered its lowest global average in a decade. V-Dem’s 2024 report identified 45 countries as autocratizing, up from 12 two decades earlier.

Three different organizations, three different methodologies, three different funding structures, three different countries of origin — and the same finding: the trend line points one direction.

The debate about where to draw the line between “liberal democracy” and “electoral democracy” is a real methodological conversation worth having. It is also a conversation that people in good faith can use as a reason to look more carefully at the underlying data — or as an excuse to look away entirely. Which response you choose says more about your relationship to the word “democracy” than about V-Dem’s methodology.

I explored how this works in Version Control for Democracy — the principle that public memory and public meaning require audit logs. When the terms change, you need to see the diff. V-Dem just showed the diff. The question is whether we read it or close the file.

What this means for your team, your family, your next meeting

You might be thinking: this is macro-level politics. What does it have to do with me?

Everything. The same mechanism runs at every scale.

“Democracy” drifts at the national level because no one pins it. “Aligned” drifts in your Monday standup for the same reason. “Ready” means three different things to three different people on your project team, and the moment you discover the gap is the moment the deadline passes. “Support” means something different to every member of your family, and the version mismatch surfaces during the crisis, when the stakes are highest and the time for clarification has already run out.

The Meaning Repair for High-Stakes Teams course teaches the mechanics of catching and repairing this kind of drift before it compounds. The four-phase model — Drift, Suppression, Repair Activation, Outcome — applies whether you’re talking about a word in a boardroom or a word in the body politic.

The V-Dem report is a Repair Activation event. A research institute spoke the first sentence. They said: by our measurement, using these criteria, this word no longer means what most Americans assume it means when they use it.

Whether that repair leads to restored shared meaning or reinforced suppression depends on what 330 million people do with the information.

It depends, in other words, on whether we treat “democracy” as a magnet word that everyone can claim without defining — or as a term precise enough to act on.

Pin it

The word: Democracy

Why it’s a magnet: It pulls toward at least five meanings — procedural, liberal, populist, tribal, and dismissive (”we’re a republic, not a democracy”) — and almost every public argument about democratic health involves people using different definitions without surfacing the difference.

V-Dem’s pin (2026): The United States retains electoral democracy (elections happen, winners take office). It has lost liberal democracy (checks on executive power, press freedom, judicial independence, civil liberties protections have fallen below the threshold that separates liberal democracies from electoral ones). The score: 0.57 out of 1.0, down from 0.8+ for three decades.

The boundary test: If your definition of democracy is “elections happen,” the U.S. still qualifies. If your definition includes “the structures between elections that prevent the concentration of unchecked power,” V-Dem says we no longer do. If your response is “we were never a democracy,” you’ve exited the conversation by rejecting the term — which is itself a meaning move worth naming. Which definition you hold determines whether the headline is alarming, irrelevant, or inadmissible — and that difference is worth surfacing before the argument starts.

One-minute script: “When you say ‘democracy,’ do you mean elections happen, or do you mean elections happen and power is constrained between elections? And if you mean ‘we’re a republic,’ are you saying representation plus rule of law — because that’s what the V-Dem report is actually measuring? Let’s agree on the definition before we argue about the score.”

Log it. Write down which definition you’re using. Notice when someone switches definitions mid-argument. That switch is the drift. Naming it is the repair.

The full evidence base for the patterns described here is in The State of Democracy: A Factual and Analytical Review of Global Trends and Tipping Points, published on SSRN. The interactive version of the team-level framework is at the-definition-gap.netlify.app. Module 1 of Meaning Repair for High-Stakes Teams is completely free — no paywall, no credit card.

Previously on What Time Binds:

• Version Control for Democracy — What happens when public memory becomes editable by power without process

• Rally Around the Flag — How crisis dynamics suppress the meaning repair a democracy needs most

• Feynman’s Real Question — Why “What do you mean?” is the most important sentence in any room

• “What Do You Mean?” as Cognitive Infrastructure — The research behind the question

• The Taxonomy: What Kind of Clarity? — Six types of meaning failure and how to spot them

Related (Medium):

• “Democracy” vs “Republic” — Testing the Aaron Russo clip against the Federalist Papers and founding texts

• “We’re a Republic” — Taking the claim at face value and asking what duties it actually imposes

Subscribe now

“Bias” is a magnet word — and this week it carried three completely different payloads into the same room. One person meant their favorite KPop member (a term of affection). Another meant unfair preference in a decision (a procedural flaw to correct). A third meant cognitive distortion (a structural problem to design around). Same six letters. Three galaxies of meaning. The room agrees to “check our bias” and immediately splits into three separate conversations.

The fix: a definition split. Write the competing definitions on the board as separate columns. Ask each person to point to the one they meant. When three people point to three different columns, the drift is visible — and you can pin the right definition for this decision.

Move: Definition Split. Why it matters: “Bias” without a column is a word doing triple duty. A 30-second split turns one loaded term into a clear operating definition.

Panel 1 — DRIFT

• Malik calls the final HallyuDayz lineup meeting: “We need to check our bias on the performer picks.”

• Juno looks up from his phone: “My bias is Jimin.”

• Rosa, holding her rubric: “We can’t have bias in the selection process. That’s the whole point of a rubric.”

• Lila catches it: “Those are three different words wearing the same outfit.”

• Pinned Terms wall: ALIGNED, READY, CLEAR, SUPPORT

• Caption: “’Bias’ carries three definitions and zero warning labels.”

Panel 2 — REPAIR

• Amina runs the reset: “Definition split. Three columns. Point to the one you meant.”

• Three columns on the board: Fandom (Identity. Joy. No fixing required.), Fairness (Procedural problem. Fix it with criteria.), Cognition (Structural. Manage it with design.)

• Rosa points to both Fairness and Cognition: “I meant both. The rubric handles B, but C is why we need someone outside the fan community to review.”

Panel 3 — INSTALL

• Lila: “Five weeks of this and I still catch myself assuming we mean the same thing.”

• The BIAS card goes up on the Pinned Terms wall: Fairness in performer selection. Rubric scores decide. One external reviewer included.

• Definition pinned with the context tag: HallyuDayz.

Move-of-the-Week: Definition Split. One more week. See this crew at HallyuDayz, March 22.

Thanks for reading What Time Binds! Subscribe for free to receive new posts and support my work.

Pinned Terms is a weekly Meaning Repair comic from What Time Binds. If this helped, share the link and tell me which word you want pinned next.

Catch up on the series: Start Here: Pinned Terms · Ep. 001: ALIGNED — Pin the term before it pins you. · Ep. 002: READY? — Scope check: includes what, excludes what? · Ep. 003: CLEAR — Verification move: say it back, say it different. · Ep. 004: SUPPORT — Five jobs in a trench coat.

▶ Watch the 3-Minute Explainer

Turn your sound on. Click PLAY. Works on any device.

In March 2026, deposition transcripts from ACLS v. McDonald went public. A man named Justin Fox (a former Tesla project manager working under DOGE) sat under oath and was asked to define “DEI.”

He couldn’t do it.

He had flagged over 1,100 NEH grants for termination using that word. A Holocaust documentary got flagged. A film about a Reconstruction-era massacre got flagged.

Over $100 million in research funding — gone — based on a term the decision-maker could not explain when the attorney asked him what it meant.

I have spent six years watching the same collapse happen in my own conversations.

On Facebook. In group chats. At kitchen tables. Someone says “DEI” or “CRT” or “systemic racism.” Everyone in the room is certain they know what it means. Almost nobody can walk you through the mechanism — step by step — of how it actually works.

The science has a name for this.

Rozenblit and Keil called it the Illusion of Explanatory Depth.

A USC Dornsife survey found 96% of Americans cannot accurately explain Critical Race Theory. Over 70% have strong opinions about it anyway.

That distance, between the feeling of understanding and actual understanding, is the definition gap.

What you’ll hear

The Deposition. Fox’s ChatGPT prompt. No definition provided to the AI. No criteria. No guardrails. A hundred million dollars in grants sorted by a word nobody pinned.

The Pattern. The same failure happening at kitchen tables. A fellow Marine argued for two years that DEI puts identity ahead of performance — then proposed exactly what a DEI program does (targeted outreach to broaden a talent pool) without recognizing it. The same person. The same conversation.

The Science. The Illusion of Explanatory Depth. People overestimate how well they understand complex systems — until you ask them to explain the mechanism step by step. Then confidence drops. Fast.

The Numbers. 96% can’t explain it. 70%+ have strong opinions anyway. That gap has a cost.

Three Repair Moves. Pin the term (”What do you mean?”). Ask for mechanism (”Walk me through how that works — step by step.”). Log the definition (”Write it down — the next conversation starts from zero unless someone saves the map.”).

Why I built it this way

This is the first time I’ve turned a What Time Binds piece into an animated explainer. I built the animated slides from scratch and deployed the whole thing as an interactive web experience with narration.

The content drove the format.

The essay I’m writing on this topic runs over 5,000 words. The explainer is a 3-minute entry point for people who want the argument, the data, and the repair moves without the full read.

The full essay — with the Facebook conversation analysis, the Cavanaugh deposition, the McDonald suppression pattern, and the complete MRCI framework — is coming soon.

The connection to the course

The three repair moves in this explainer come directly from Module 1 of Meaning Repair for High-Stakes Teams. The Meaning Risk Snapshot — the diagnostic tool that measures where your team is most vulnerable to this kind of drift — is the first deliverable in the course.

If the explainer lands, the course gives you the full operating system.

Module 1 is free. Always. Start here.

Share the explainer link — the-definition-gap.netlify.app — with anyone who needs to hear: “What do you mean?”

Gosh, this is a long one.

Friday evening. I shared a meme on Facebook.

The meme was sharp. Rick Wilson, pointing out that Russia is giving Iran intelligence to target American forces while Ukraine is offering to defend them. The rhetorical question: “So tell me again — which one is the ‘enemy,’ and which one is the ally?”

I posted it with five words: “Waiting for the mental gymnastics.”

I knew what was coming. I’ve been on the internet long enough.

By the next day, I had two very different responses.

One moved toward clarity. One didn’t. I deleted the one that didn’t. Then the real conversation started — and it turned into the most honest public exchange about political identity I’ve been part of in years.

This essay is about what happened in that thread, what the research says about why it happened, and what it taught me about when to protect the conversation and when to protect yourself from it.

Subscribe now

The thesis, up front

Here it is: Deleting a comment and defriending someone can be an act of cognitive infrastructure maintenance. You’re clearing the space so the harder, slower, more productive conversation has room to breathe. The research supports this. And my own experience this week proved it in real time.

I’m also updating a position I took publicly. In my earlier essay, Protect Your Attention Like Critical Infrastructure, I wrote: “I’m not defriending people. I’m unfollowing their content. That’s a key distinction.”

That was the position. This week, the position evolved — because someone’s conduct made even unfollowing insufficient. The person wasn’t holding a different view. They were making the kind of comment that degrades the conditions under which productive disagreement is possible.

I’ll walk you through what happened. Then I’ll show you what the research says about the forces at work. Then I’ll give you something you can use.

Two responses, two outcomes

The first response came from a friend who leans right — libertarian, guided by his Christian faith. He and I disagree on a lot. He asked a question that, on the surface, looked like pushback: “I’m curious to who has ever suggested that Russia was our Ally?”

I did what I teach. I asked: “What do you mean?”

He clarified. The meme, he said, implies that MAGA believes Russia is an ally and Ukraine is an enemy. His understanding was that MAGA believes neither are allies.

Fair enough. Now I knew what he actually meant, and I could respond to what he said — with data and sources, not assumptions. I shared the Gallup finding that 40% of Republicans consider Russia an ally or friendly, compared to 25% of Democrats. The Pew data showing Republican favorable views of Russia doubling. The Vanderbilt Unity Poll finding that 52% of MAGA-identifying Republicans said Putin is a better president than Biden. Trump calling Putin’s invasion “genius” on national radio. Republican favorable views of Zelenskyy collapsing from 53% to 28% in a single year.

My friend took it in. Then he said something remarkable:

“I guess I mostly don’t understand the term MAGA anymore.”

Read that again. A person who leans right, who would naturally resonate with the aspiration of making America great, just told me the word has drifted beyond his reach. The label no longer maps onto what he actually believes.

That’s meaning drift, happening in real time, in a Facebook comment thread.

I walked him through the mechanism. Words carry more than what we intend them to carry. They accumulate meaning through how they get used, who uses them, and what gets done under their banner. “MAGA” started as a slogan. It became the name of a political movement with specific leaders, specific rhetoric, and specific policy positions. Once that happens, the word starts carrying all of that weight, whether any individual signed up for all of it or not.

He landed on an insight that George Carlin articulated years ago (and I’m paraphrasing his paraphrase here): when your ideology becomes your identity, a disagreement stops being “I think your ideas are wrong” and becomes “I think your existence is wrong.” Criticism stops being a critique of ideas and becomes an attack on your being.

I told him: it’s a human issue. And it can be repaired.

That exchange — the whole thing, in public — is what the MRCI framework calls a completed repair sequence. Drift detected. Repair activated. Shared understanding reached. Relationship intact. The conversation continued.

Now the second response.

Someone else left a comment. I won’t name them. I deleted it as soon as I read it. It wasn’t moving the conversation toward mutual understanding based on verifiable evidence. It was the kind of comment that, if left standing, would have changed the temperature of the entire thread.

I messaged my friend privately to let him know. The comment had been directed at him. I told him I’d deleted it, and that if he wanted to, he could reach out to the person directly — their relationship was theirs to manage. I was protecting my space, not controlling his choices.

He hadn’t seen the comment. He’d been on a date with his partner and had put Facebook away for a few hours.

He responded, characteristically, with Scripture. He quoted 1 Peter 3:15-16, about entering discussions “with gentleness and respect, so that when someone walks away from the conversation, they can’t speak poorly of you without feeling shame.” He applied it beyond faith to all discourse: “I can disagree with you without being disrespectful or attacking you personally... and when we walk away from that conversation, we still hold a mutual respect for one another and the relationship remains intact.”

I told him I agreed. My entire doctorate was built around doing exactly that.

Here’s what I need you to see: the deletion created the conditions for that private exchange. If the hostile comment had stayed up, the thread would have shifted from meaning repair to damage control. The two of us would have been managing someone else’s conduct instead of doing the harder, better work of understanding each other.

Deleting the comment was triage, not retreat.

What I said before — and why I’m updating it

In February, I published Protect Your Attention Like Critical Infrastructure on this Substack. The core argument was that attention moves meaning the way roads move goods and power grids move energy. When that system gets congested, contaminated, or hijacked, everything downstream suffers — your judgment, your patience, your relationships, your ability to plan, your ability to learn.

I laid out a specific position: “I’m not defriending people. I’m unfollowing their content. That’s a key distinction. I can value the person, keep the relationship channel open, and still refuse to let an engagement-optimized stream rent space in my head.”

I also committed to building “structured cross-traffic” — following a small number of people I disagree with who show intellectual honesty. People who cite sources, correct themselves, and avoid dehumanizing language.

That framework still holds. My friend is exactly the kind of person I was describing. We disagree. He shows intellectual honesty. He engages with evidence. The conversation continues.

The person whose comment I deleted? They failed that test. The relationship channel was open. The intellectual honesty wasn’t there. And when someone’s conduct — verbal attacks, not ideas — makes the space unsafe for the people who are engaging honestly, unfollowing isn’t enough. The comment needed to go. And after reflection, so did the connection.

This is an honest update, and I want you to see me making it in public. Positions should evolve when the evidence changes. The evidence this week was that maintaining a connection with someone whose conduct degrades your shared spaces isn’t worth the trouble.

“MAGA” is a magnet word. “Civility” is its shadow.

His admission — “I don’t really understand the term MAGA anymore” — is the essay in miniature. A word that once felt like it pointed toward his values had accumulated so much additional meaning that it no longer mapped onto what he actually believes.

In the MRCI framework, this is what a magnet word does. It pulls everyone toward apparent agreement while meaning something different to each person. Everyone nods. Everyone scatters.

The research on this is staggering.

Yang et al. analyzed broadcast transcripts from CNN and Fox News between 2010 and 2020 and found a 112% increase in semantic polarity across politically charged keywords starting in 2016. Terms like “racist,” “police,” “immigrant,” and “health care” are used in measurably divergent ways by the two networks — and this media divergence predicts social media polarization downstream.

Gentzkow, Shapiro, and Taddy studied Congressional speech from 1873 to 2016 and found that partisan language diverged sharply in the early 1990s after remaining low and constant for over a century. Their measure captures the ease with which an observer could guess a congressperson’s party from a single sentence. Democrats say “estate taxes” and “undocumented workers.” Republicans say “death taxes” and “illegal aliens.” Same referent. Incommensurable frames.

George Lakoff’s work explains why this happens. He argues that American political thinking is structured by an unconscious metaphor of the nation as a family. The “Strict Father” model — conservative — values self-discipline, self-reliance, and moral authority. Freedom means freedom from external interference. The “Nurturant Parent” model — progressive — values empathy, care, and community support. Freedom means freedom to pursue one’s potential. When both sides say “freedom,” they activate entirely different conceptual systems.

And here’s the finding that should stop you: Nakwon et al. combined word embeddings with fMRI neuroimaging and found that the same political words activate different brain patterns in people at different points on the ideological spectrum. Magnet words are measurable. The divergence is neurological.

“MAGA” has undergone exactly this kind of splintering. For my friend, the aspiration — a thriving, great America — remained constant. For the movement that claimed the phrase, the word now indexes a specific set of leaders, rhetoric, and positions. The 52% who say Putin is a better president than Biden aren’t using the word the way he uses it. They’ve moved the word somewhere he didn’t go.

Now layer on “civility,” which functions as the shadow magnet word underneath every defriending decision.

The Georgetown Institute of Politics and Public Service found that 90% of voters are concerned about “uncivil and rude behavior of politicians.” Bipartisan consensus, right? Except they fundamentally disagree on who is being uncivil. Republicans blame Democratic leaders and CNN/MSNBC. Democrats blame Republican leaders, Fox News, and Trump. On a 0-100 scale measuring proximity to civil war, the average response was 67.23.

Ninety percent of Americans share the word “civility.” They assign it mutually exclusive referents.

Patrícia Rossini’s research provides the analytical scalpel. She argues that “incivility” — violations of respectful communication norms like hostile tone and mockery — must be distinguished from “intolerance” — attacks that undermine democratic values like racism, threats, and dehumanization. Her finding: incivility actually correlates with meaningful discursive engagement. People argue rudely when they care. Intolerance shows up in homogeneous discussions targeting minorities, where it does real democratic damage.

He was engaging in productive friction. His initial question pushed back on my framing. That’s incivility in Rossini’s technical sense — and it’s healthy. The comment I deleted was something else.

Teresa Bejan’s Mere Civility traces this ambiguity to three competing historical conceptions: suppress disagreement for order (Hobbes), exclude the “uncivil” for concord (Locke), or practice a minimal, grudging tolerance compatible with deep disapproval (Roger Williams). The problem — the one that lands in your Facebook thread at 9 PM on a Friday — is that when someone invokes “civility,” they rarely specify which version they mean.

The MRCI framework offers a way through. Pin the word. In this room, “civility” means: you can disagree sharply, cite sources, push back on my framing, and tell me I’m wrong — as long as the conversation can continue. It excludes: personal attacks, dehumanizing language, and comments that exist to inflame rather than inform. It includes: awkwardness, discomfort, and the possibility that you’ll change your mind. We’ll revisit when the conditions change.

That’s what I was enforcing in my thread. I just didn’t have the language for it at the time.

The attention science: why the delete was a leadership decision

Here’s where this connects to your work life, your decision-making, and the bandwidth you’re trying to protect.

Sophie Leroy’s attention residue research (published in Organizational Behavior and Human Decision Processes) demonstrates that when people switch tasks, part of their attention stays with the prior task. If that task was unfinished or emotionally charged — like a hostile political exchange on your feed — the residue is stronger and subsequent task performance suffers significantly. Every hostile encounter you leave standing in your feed creates cognitive drag on everything you do next.

Mullainathan and Shafir’s work on scarcity shows that depletion of any resource — money, time, or attention — imposes a cognitive tax equivalent to 13-14 IQ points. That moves someone from “average” to “borderline deficient.” This applies directly to attention scarcity from political information overload. The bandwidth tax is real, it’s measurable, and it degrades your capacity for the decisions that matter most.

Hughes et al. found in a time-lagged daily diary study that doomscrolling at work increases rumination, which directly decreases work engagement. The more neurotic you are, the worse the effect.

And Smith, Hibbing, and Hibbing surveyed 800 Americans and found that roughly 40% reported politics as a cause of stress, 20% reported losing sleep or feeling depressed over politics, and 4% reported suicidal thoughts related to political engagement. Extrapolated nationally, that last figure represents roughly 10 million adults. The APA’s “Stress in America” surveys show the trend accelerating: 57% in 2017, 68% in 2020, 74% in 2024 said politics was a significant stressor.

When I deleted that comment, I wasn’t being thin-skinned. I was doing what any good infrastructure manager does: identifying a contaminant in the system and removing it before it degraded the downstream output.

The downstream output, in this case, was the conversation that mattered.

The echo chamber objection — and why it doesn’t apply here

The most common pushback on defriending is that it creates echo chambers. “You’re just surrounding yourself with people who agree with you.”

Two responses.

First, the data. Bail et al. published a study in the Proceedings of the National Academy of Sciences where they had Republicans follow a liberal Twitter bot for a month. The result? They became substantially more conservative. Mere exposure to opposing views, in a social media context, can increase polarization. The quality of the encounter matters more than its existence. Removing a source of contemptuous engagement may actually reduce the polarization cycle.

Second, the distinction that the Markkula Center for Applied Ethics at Santa Clara University draws cleanly: “I won’t unfriend or block people simply on the basis of whom they support, or why. I will do it based on how they actually engage in dialogue.”

I didn’t remove a viewpoint from my thread. I removed conduct that was degrading the conditions for productive disagreement. My friend’s viewpoint — right-leaning, libertarian, faith-guided, skeptical of labels — stayed. And it made the thread better.

Alexis Elder, writing in Ethics and Information Technology, argues from an Aristotelian account of friendship that unfriending tools can actually promote better civic conversation by allowing people to remove bad-faith actors while maintaining healthy diversity.

The question worth asking is the one a researcher proposed in the Journal of Deliberative Democracy: the health of a discourse should be measured by a single question — “Will the conversation continue?”

When someone is verbally attacking others and you can’t ask “What do you mean?” without triggering more aggression, the conditions for meaning repair have collapsed. At that point, deleting the comment doesn’t end the conversation. It recognizes that the conversation was already over — and protects the one that’s still alive.

What this looks like in MRCI terms

For those of you following the Meaning Repair for High-Stakes Teams course, here’s the framework map of what happened in my thread:

My friend’s path — the completed repair sequence:

Phase 1: Drift. We both read the meme. We assigned different meanings to “MAGA” and different functions to the meme itself (he read it as a factual claim; I read it as rhetorical irony). Meaning diverged silently.

Phase 2: Suppression — avoided. He asked his question publicly. He didn’t swallow it. He didn’t snark. He didn’t DM me something passive-aggressive. He put the confusion on the table. That’s the hardest move, and he made it.

Phase 3: Repair Activation. I asked “What do you mean?” He clarified. I responded with evidence and pinned the term: here’s what “MAGA” means in polling data, in leadership rhetoric, and in policy positions. Here’s the gap between the aspiration and the movement. He processed it and arrived at his own insight — ideology becoming identity. Meaning restored.

Phase 4: Outcome. The relationship is intact. The conversation is on the record. Future-us inherits the map. He knows where I stand. I know where he stands. We both know where the word fractured.

The deleted comment — the collapsed sequence:

Phase 1: Drift. Same meme. Same divergence.

Phase 2: Suppression — inverted. The person didn’t suppress repair. They suppressed the conditions for repair. The comment’s function was to attack, not to understand. When your move makes it impossible for anyone else to ask “What do you mean?” safely, you’ve collapsed the sequence before it can start.

Phase 3: Repair Activation — blocked. No repair was possible. You can’t pin a word with someone who’s throwing words as weapons.

Phase 4: Outcome. I deleted the comment. I removed the connection. The thread survived. The better conversation happened.

A repair protocol for your own feed

Here’s what you can install from this.

Before you delete or defriend, run the test:

“Can I ask ‘What do you mean?’ and get an answer that moves us toward understanding?”

If yes — even if the answer is uncomfortable, even if the person is wrong, even if the conversation is hard — stay in it. That’s the repair path.

If no — if asking the question would trigger escalation, personal attack, or more heat with no light — you’re looking at conduct that has collapsed the conditions for repair. Removing it is triage.

Three scripts for engaging across difference (the repair path):

“When you say [term], what do you mean by that? I want to make sure I’m responding to what you actually think, not what I assume you think.”

“I hear the word [term] carrying a lot of weight right now. Can we pin down what it means for this conversation?”

“You and I might be using [term] differently. Here’s what I mean by it — [definition]. Does that match your understanding?”

Three questions before you sever (the triage path):

Is this person attacking ideas, or attacking people? (Rossini’s incivility-vs.-intolerance line.)

If I stay connected, does their presence degrade my ability to engage productively with others? (The attention residue question.)

Am I severing because I’m uncomfortable with their position, or because I’m unable to engage with their conduct? (The Markkula Center’s position-vs.-conduct distinction.)

After you sever:

Log what happened. Write it down — even privately. Not to justify yourself, but because the next time you face this decision, you’ll have a map. Future-you inherits what present-you records. That’s time-binding.

The word is telling you something

Let me return to where we started.

The meme is still on my page. The deleted comment is gone. The conversation my friend started is still visible — and it’s one of the most substantive public exchanges I’ve had on Facebook in years.

He told me the word “MAGA” no longer maps onto what he actually believes. That discomfort — the sense that a word moved without your permission — is the word telling you something. It’s meaning drift made personal.

The aspiration didn’t change. The movement changed. And the word followed the movement, not the aspiration.

I told him: it’s a human issue. Both sides of the aisle do it. When ideology becomes identity, every disagreement becomes an existential threat, and the space for repair disappears.

But here’s what I believe, and what this newsletter is built on: the space can be rebuilt. One question at a time. One pinned definition at a time. One decision to protect the conversation instead of winning it.

I deleted a comment this week. It’s the first time I’ve done that on a political post. I also defriended the person who wrote it.

I’m not proud of it. I’m not ashamed of it. I’m clear about why I did it.

The conversation that mattered was still happening. I chose to protect it.

What word is causing drift in your world right now? And can you still ask “What do you mean?” — or has the space for that question already collapsed?

Hit reply. I read everything.

This essay is part of the What Time Binds series on meaning, attention, and what we pass forward. It connects to the Meaning Repair for High-Stakes Teams course — where the MRCI framework (Drift → Suppression → Repair Activation → Outcome) is taught as a repeatable practice for any team, family, or community that needs to coordinate under pressure. Module 1 is completely free at what-time-binds.com.

If you’re reading this and thinking about your own feed, your own deleted comments, your own severed connections — you’re already doing the work. The fact that you’re asking whether the decision was right means you haven’t stopped caring about the relationship. That matters.

You completed the Meaning Risk Snapshot. You scored your team. You named some risks. Now, the question that determines whether any of this sticks.

What the next ten weeks actually feel like

You will try a repair move in a meeting, and it will come out wrong. The sentence you practiced at your desk will arrive in the room with an extra clause and a nervous qualifier tacked on the end. Someone will give you a look. You will wonder if the move made things more awkward, not less.

This is the correct experience. It means the learning is working.

Research on habit formation is consistent on one point: health-related habits take two to five months to develop, with substantial variation between individuals. Starting slowly and increasing gradually, what we’ll call habit laddering, maintains new practices 1.5 times longer than going all-in from day one. A ten-week course falls inside that window. But it means you are still mid-construction at Module 3, Module 5, and probably Module 8.

The awkwardness is not a sign that you’re failing. It is the felt experience of building a new default. The old default (swallow the doubt, let the ambiguity slide, assume everyone understood the same thing) took years to install. The new one needs time and repetition under real conditions.

So: expect the wobble. Practice the move anyway. The goal for the first three modules is frequency, not fluency. Use the repair stance once this week. Use it badly if necessary. The reps compound.

Subscribe now

What your Meaning Risk Snapshot is telling you

You now have a scored baseline. A set of numbers attached to specific communication risks on your team. Some of those numbers confirmed what you already suspected. Others may have surprised you.

Here is what to do with the snapshot before Module 2 arrives.

Look for the cluster. Most teams don’t have ten equally severe risks. They have two or three failure modes that account for the majority of their coordination problems. Which items scored lowest? Where did the numbers cluster? That cluster is your team’s drift signature, the pattern of meaning failure that repeats across meetings, handoffs, and decisions.

Notice what you couldn’t score. Some items on the snapshot may have felt unanswerable. You weren’t sure how your team handles ambiguity because you’ve never watched for it with this lens. That uncertainty is data. It tells you where your team’s communication patterns are operating below conscious awareness, which is exactly where drift lives.

Hold the cultural context field. If you filled in the Meyer dimensions (power distance, communication directness, comfort with silence), sit with those ratings for a moment. They shape which repair moves will land easily on your team and which ones will need adaptation. A team with high power distance will need different activation pathways than a team where anyone can interrupt anyone. We’ll build those pathways starting in Module 4.

You’ll retake this snapshot at Day 30 and Day 60. The numbers will move. The question is whether they move because your team’s communication actually changed, or because your awareness sharpened. Both matter.

Transfer Bridge

The Meaning Risk Snapshot was designed for work teams. But meaning drifts everywhere when people coordinate under pressure, with incomplete information, or with competing assumptions.

This week’s prompt: Where outside your current work role would this diagnostic have changed an outcome recently?

A few places to look:

A medical appointment where you and the provider were using the same words to describe different things. A family conversation about plans, where “we agreed” turned out to mean two different commitments. A negotiation, with a contractor, a landlord, a co-parent, where the key term was never pinned down, and the ambiguity surfaced later as conflict.

Pick one. Run the snapshot against it mentally. Which failure type would you flag? Where was the drift? When did you sense it? What kept you from naming it?

That last question, what kept you from naming it, is the one this entire course circles around.

Discussion

Three questions for the comments. Pick the one that pulls you.

1. What is your team’s drift signature? Which two or three risk areas clustered at the bottom of your snapshot, and does that match what you would have guessed before scoring?

2. When you think about the cultural context field, what is the single dimension (power distance, directness, silence comfort) that most shapes how your team handles ambiguity? What does that look like in practice?

3. For the Transfer Bridge: describe one non-work conversation where meaning drifted, and you noticed but didn’t repair it. What was the cost?

A standing invitation. If your example from this week — whether the repair move worked, went sideways, or revealed something you didn’t expect — would be useful as a case study for future course material, share it in the comments with the tag [CASE STUDY]. I collect these (with permission, anonymized) to strengthen the evidence base behind this framework. The more situated examples we accumulate from real teams and real conversations, the sharper the tools become for everyone who follows.

Next: Module 2 — The Taxonomy of Meaning Failures. You spotted the risks. Now you learn to name exactly what kind of failure you’re looking at. When you can classify a breakdown in real time, the repair move becomes obvious.

This is Module 1, Post 3 of Meaning Repair for High-Stakes Teams, a 10-module course on the What Time Binds Substack. New to the course? Start with the Meaning Repair Lexicon and Module 1, Post 1: Why Meaning Fails Silently.

The tools for diagnosing communication breakdowns exist. A team assessment instrument validated across 360 professionals in nine health systems (Cronbach’s α = 0.97) measures exactly the teamwork dimensions where meaning drifts. This post turns that research into something you can use this week.

Why a snapshot, and why now

In Post 1, you met the four phases and the 100%-vs.-12% gap. The framework makes conceptual sense. But concepts don’t repair meaning. Diagnosis does.

The Meaning Risk Snapshot is your team’s baseline. It tells you where meaning is most likely to fail, how your team’s culture shapes that failure, and what to measure again at Day 30 and Day 60 so you can track whether the moves in this course are landing.

D1 Meaning Risk Snapshot Fillable

46.8KB ∙ PDF file

Download

Download

Here’s how to fill one out.

The three parts of the snapshot

The snapshot has three sections, each about five minutes. Fifteen minutes total. You get a quantitative score you can track over time, a cultural context map that shapes how you adapt repair moves in later modules, and a narrative that names your specific targets.

Part 1: Meaning Drift Risk Score — ten Likert-scaled items adapted from validated instruments. This gives you a number.

Part 2: Cultural Context Map — a self-assessment based on Erin Meyer’s eight cultural dimensions. This gives you a picture of where gaps between team members might amplify drift

Example Part 2 Cultural Context Map

68KB ∙ PDF file

Download

Download

.Part 3: Narrative Risk Summary — three open questions. This gives you language.

Part 1: The Meaning Drift Risk Score

These ten items draw on the Team Assessment Tool (Ali et al., 2024), validated across nine U.S. health systems with exceptional reliability, and the Resuscitation-Specific Organizational Culture Instrument (Handley et al., 2024), which measures role clarity, shared mental models, closed-loop communication, team adaptability, and psychological safety. I’ve adapted their clinical language into terms that work for any team—project, operational, family, or community.

Score each item from 1 (strongly disagree) to 5 (strongly agree) based on your team’s typical behavior, not its best day.

1. Shared understanding of terms. When our team uses key terms (project names, status labels, priority levels, role titles), members mean the same thing.

2. Closed-loop communication. When someone makes a request or gives an update, the receiver confirms their understanding back to the sender.

3. Role clarity under pressure. During high-stakes or time-pressured work, every team member knows who is responsible for what.

4. Handoff quality. When work, decisions, or information transfer from one person or shift to another, the receiving person gets enough context to act without guessing.

5. Willingness to voice concern. Team members speak up when they notice something that seems wrong, incomplete, or unclear—even if the person they’d be questioning is more senior.

6. Leader receptivity. When someone raises a concern or questions a decision, the team’s leader(s) respond in a way that encourages continued speaking up.

7. Assumption surfacing. Our team regularly checks whether people are working from the same assumptions about scope, timeline, priorities, or definitions.

8. Debrief practice. After significant events (launches, incidents, major decisions, difficult conversations), our team reviews what happened and what to change.

9. Adaptability when plans change. When new information arrives mid-task, our team can redistribute roles and update shared understanding quickly.

10. Context travel. Information that is clear in one setting (a meeting, a document, a conversation) reliably reaches the people and settings where it’s needed next.

Scoring: Add your ten responses. Range is 10–50.

• 40–50 (Strong grounding): Your risks concentrate where pressure compresses the acceptance phase—handoffs, escalations, fast-moving situations. Focus on precision protocols (Modules 5–7).

• 25–39 (Mixed grounding): Some rhythms work; others drift. Most teams sit here. The narrative section (Part 3) helps you identify which items pull your score down.

• 10–24 (Significant drift risk): Meaning failures are generating rework and frustration on a recurring basis. The suppression-phase work (Modules 3–4) and cultural context map (Part 2) are especially important—the barriers are likely structural, not skill-based alone.

Record your score. You’ll retake this at Day 30 and Day 60.

Personal Repair History (unscored)

Before you move to the Cultural Context Map, take sixty seconds on two questions. These aren’t scored. They’re for your awareness only.

The last time it went badly. Think about the most recent time you raised a concern, asked for clarification, or questioned a decision, and it didn’t go well. Maybe you were dismissed. Maybe the conversation got uncomfortable. Maybe nothing changed. How long ago was that? Was it on this team or a different one?

The feeling that shows up first. When you imagine asking “what do you mean by that?” in your next meeting, what’s the first feeling that arrives: curiosity, caution, or dread?

That feeling is data about your own repair threshold. It was shaped by every team you’ve been on, every organization you’ve worked in, every time you spoke up and got rewarded or punished for it. Your threshold travels with you. It doesn’t reset when you change jobs. A person who has been dismissed for raising concerns across three organizations will experience “what do you mean?” differently than someone whose teams have always welcomed the question, even if both people are now sitting on the same team with the same manager.

This course can lower your threshold over ten weeks. But it helps to know where it sits right now, so you can feel it moving.

Part 2: The Cultural Context Map

Erin Meyer’s research produced a key insight: what matters for communication is the gap between team members on any dimension, not where any single person falls. A team where everyone communicates indirectly has fewer drift risks than a team split between direct and indirect communicators. The distance creates the danger.

For each of eight dimensions, place your team on the spectrum. If your team has significant spread, note the range rather than forcing a single point. The full snapshot instrument includes all eight with meaning-repair implications for each. Here are the dimensions:

1. Communication style: Low-context (explicit, precise) ←→ High-context (implicit, read-between-the-lines)

2. Feedback approach: Direct negative feedback ←→ Indirect negative feedback (softened, private)

3. Leadership expectation: Egalitarian (challenge-the-boss is normal) ←→ Hierarchical (deference to rank)

4. Decision-making: Consensual (discuss until agreement) ←→ Top-down (leader decides after input)

5. Trust basis: Task-based (competence and reliability) ←→ Relationship-based (personal connection over time)

6. Disagreement style: Confrontational (open debate is productive) ←→ Avoidant (confrontation damages relationships)

7. Scheduling orientation: Linear-time (sequential, deadline-driven) ←→ Flexible-time (adaptive, fluid)

8. Comfort with silence: Low (3 seconds feels long) ←→ High (12+ seconds is normal)

What to look for: Circle any dimension where your team spans more than half the spectrum. These are your cultural gap zones. In Module 4, you’ll use this map to decide whether leader-initiated repair, anonymous concern channels, or written pre-work would serve your team better than default verbal protocols.

One note: you’re rating your team as you see it, which means your placement reflects your own cultural lens. The goal isn’t objective measurement. It’s awareness of distance—enough to ask, later, “Would this move work the same way for everyone on my team?”

Quick inventory: where does your team’s meaning travel?

List the three channels where your team’s most consequential conversations happen. These might be live meetings, Slack threads, email chains, shared documents, presentations, or phone calls.

For each one, note whether it allows real-time correction, meaning the other person can immediately say “wait, that’s not what I meant” and you can adjust on the spot.

The channels that allow real-time correction are where repair happens naturally. The channels that don’t (a presentation deck, a policy memo, a recorded announcement, a published article) are where drift compounds fastest, because the acceptance phase has no space to occur.

This matters because teams routinely move their most important conversations from live channels to one-way channels without noticing what gets lost. A decision made in a meeting gets documented in an email. An email gets summarized in a slide. A slide gets presented to a room that has no context for the original conversation. At each step, the opportunity for someone to say “wait — what do we mean by that?” shrinks.

When you see your team moving a consequential conversation from a live channel to a one-way channel, that’s a moment to pause and ask: has everyone confirmed they mean the same thing, or are we locking in a version of shared understanding that was never actually verified?

Part 3: The Narrative Risk Summary

The score tells you how much drift you have. The map tells you the cultural terrain. This section names the specific failure points.

Answer three questions in writing. A few sentences each is enough.

1. The recurring failure. Describe one communication breakdown that has happened more than once on your team. Where does meaning typically diverge? When do people usually discover the gap?

2. The suppressed concern. Think of a time in the last month when you (or someone you observed) noticed confusion or disagreement but did not raise it. What was the context? What made speaking up feel difficult or pointless?

3. The costliest gap. What is the single communication pattern that costs your team the most—in rework, missed deadlines, strained relationships, or poor decisions? Name it as specifically as you can.

These three answers become the raw material for the rest of the course. The recurring failure is where you’ll practice spotting drift (Module 2). The suppressed concern is where you’ll practice repair stances (Module 3). The costliest gap is what your adoption plan (Module 10) should target first.

Your practice rep

Complete the full Meaning Risk Snapshot for your primary team this week. “Primary team” means the group of people whose communication quality most affects your work or life right now—your direct team at work, a project group, a family decision-making unit, a board you serve on.

Fill in all three parts. Record the date and your Drift Risk Score. File it somewhere you’ll find it at Day 30.

If you’re working through this course with a team, have each member complete the snapshot independently before comparing results. The places where your scores diverge are, themselves, meaning drift—and they’re some of the most useful data you’ll generate.

What's next: Post 3 — The Ten-Week Build. Your snapshot gave you the baseline. The next post sets the habit formation expectation for the course and introduces the Transfer Bridge — a prompt that moves repair practice beyond your primary team. Then in Module 2, you'll learn to classify the failures you named in Part 3. The taxonomy of meaning failures gives you the vocabulary to spot drift as it happens, before the cost arrives.

This is Module 1, Post 2 of Meaning Repair for High-Stakes Teams, a 10-module course on the What Time Binds Substack. The complete Meaning Repair Lexicon defines every term used in this course.

“Support” is a magnet word. It sounds like a commitment, but it’s actually a blank check that five people fill in differently. One person hears “I’m handling logistics.” Another hears “I’m promoting it.” A third hears “I thought we were just showing up.” The room agrees. The work scatters.

The fix: a role map — ask each person to write one sentence defining their version of “support.” When five cards go up, and every card says something different, you’ve just made the invisible org chart visible. Now you can assign it before the week assigns it for you.

Move: Role Map. Why it matters: “Support” without decomposition is an entire org chart hiding inside one word. A two-minute role map turns a vague commitment into named responsibilities.

Panel 1 — DRIFT

• Malik announces at the whiteboard: “Big news — we’re supporting the HallyuDayz fan event next week!”

• The HallyuDayz flyer (March 22) is pinned to the board. The Role Map is already filling in — but everyone wrote a different job.

• Pinned Terms wall visible: ALIGNED, READY, CLEAR

• Caption: “’Support’ is a magnet word. Everyone hears their own job description.”

Panel 2 — REPAIR

• Amina runs the reset: “Quick reset. Everyone write one sentence: what does ‘support’ mean to you for this event?”

• Five roles surface: Promo (Malik), Logistics (Rosa), Attendance (Juno), Program (Lila), Facilitation (Amina).

• The SUPPORT card goes up on the Pinned Terms wall — formatted as a Role Map with all five names and responsibilities.

• Hart holds the Decision Log. Logged.

Move-of-the-Week: Role Map. See this crew at HallyuDayz, March 22.

Thanks for reading What Time Binds! Subscribe for free to receive new posts and support my work.

Pinned Terms is a weekly Meaning Repair comic from What Time Binds. If this helped, share the link and tell me which word you want pinned next.

Catch up on the series: Start Here: Pinned Terms · Ep. 001: ALIGNED — Pin the term before it pins you. · Ep. 002: READY? — Scope check: includes what, excludes what? · Ep. 003: CLEAR — Verification move: say it back, say it different.

Above: Wooden Scrabble tiles spell ORDER in a neat horizontal row against a white surface. Below them, tiles that could spell CHAOS are scattered loosely — some face-up, some tilted, none assembled. The image captures the essay's central argument: ORDER and CHAOS share the same alphabet. What separates them is whether the pieces are held in a shared arrangement — and who gets to decide what that arrangement means.

Subscribe now

Brian was my minister growing up. I’ve known him since I was a kid — the kind of pastor who reads the original Greek, argues theology like it matters, and does not confuse passion for rigor. His page goes by “Out of the Box Pastor,” which is apt. He’s the type of theologian who treats sloppy doctrine the same way a good editor treats sloppy prose: with a red pen and zero patience.

The post was a graphic. Dark red background. White text.

“Trump is not anointed to start Armageddon, so Jesus can return. That eschatological nonsense is 100% extra-biblical fiction.”

I commented. Told him I’d been researching exactly this — that framing matching that description had been appearing in military briefings, and that I was probably going to write about it on my Substack.

What came back in the thread stopped me.

Brian called the theology wrong. Then he placed it. He named the New Apostolic Reformation — the NAR — as the likely engine behind what was showing up in the briefings. He described the movement as “highly sectarian,” as promoting “the use of violence in order to dominate.” He called what was happening “ridiculously dangerous — two ultra right-wing radical fundamentalist ideologies at war with each other.”

I’d already found the HuffPost piece in my research — the same story Brian was describing, documented at more than 30 installations. I dropped the link in the thread. Brian’s response: “within the military noncoms, from what I gather.”

A minister on Facebook had mapped the problem more precisely than most of the institutional analysis I’d read.

I kept thinking about that. About what it means that Brian could see it clearly from the outside. And what it says about what’s happening on the inside.

Here’s what I think it means. The word at the center of this — the one holding the entire structure together — has changed its meaning in ways that make questioning it nearly impossible through normal channels.

That word is ORDER.

The word that runs everything

In the United States military, order is the structural foundation of everything.

The Uniform Code of Military Justice — the UCMJ — makes obedience to lawful orders a legal obligation. Articles 90, 91, and 92 create a presumption of lawfulness: a service member who refuses an order does so at their own peril. Disobedience is punishable by court-martial.

When we talk about order in this institutional sense, we mean a directive derived from constitutional authority — from the President as commander-in-chief, confirmed by Congress, constrained by the laws of armed conflict, subject to review by judge advocates. That authority is secular and bounded. It can be checked against law. A service member can assess it. A military attorney can evaluate it. An inspector general, at least in theory, can investigate it.

That definition has a referent. It lives in a verifiable system.

Now listen to what commanders at more than 50 military installations have reportedly been telling their troops, according to complaints filed with the Military Religious Freedom Foundation earlier this year:

The Iran war is part of God’s plan. Trump was anointed by Jesus to light the signal fire in Iran to cause Armageddon and mark his return to Earth. This is what we are called to do.

Same word. Same rank structure. Same legal obligation to obey.

Different operating system entirely.

What happens when a meaning regime swaps

The word order didn’t change. The chain of command didn’t change. The UCMJ didn’t change.

What changed was the source of authority behind the word.

A lawful order derives its authority from constitutional structure — secular, verifiable, bounded. A divinely ordained order derives its authority from God’s will. That authority cannot be reviewed. It cannot be appealed. Questioning it becomes an act of apostasy.

When a commander frames a combat operation as part of biblical prophecy, the word order is still in the room. The rank structure is still in the room. The legal obligation to obey is still in the room. But the meaning regime behind the word has shifted from a system designed for human review to one that, by definition, forecloses review.

This is what I mean when I call order a magnet word in this context: it’s a term carrying institutional weight in one register and theological weight in another, and the two do not operate on the same rules. Most magnet word failures are recoverable — people walk out of the meeting with different pictures of “finalize” or “priority,” and you can repair it when the collision surfaces. This failure mode is different. The UCMJ compels obedience to the institutional frame. The theological framing makes questioning the transcendent frame an act of spiritual betrayal. A service member is caught between a legal obligation to follow and a spiritual cost for asking.

The order, on paper, may be completely lawful. What has been done to it does not show up on paper at all.

How the pattern built

This is not a theoretical concern.

In March 2026, the Military Religious Freedom Foundation reported receiving more than 200 complaints from service members at more than 50 installations across every branch. The complaints described commanders framing preparations for operations against Iran in explicitly eschatological terms — language connecting military action to end-times prophecy and the physical return of Jesus Christ.

The MRFF was founded by Michael Weinstein, a 1977 Air Force Academy graduate, former JAG officer, and Reagan White House legal counsel. It has represented nearly 100,000 military personnel — 95% of whom identify as practicing Christians — and has been nominated for the Nobel Peace Prize three times.

And those complaints do not arrive in isolation. They arrive inside a pattern that has been building for at least the past year.

In May 2025, monthly Christian worship services began at the Pentagon, held during working hours, livestreamed on the Department of Defense’s internal television network. Secretary Hegseth, leading one of the services: “This is precisely where I need to be... in prayer, on bended knee, recognizing providence of our Lord and Savior Jesus Christ.” The DoD official account posted that same day: “We have gathered at Pentagon for our monthly worship service. We are One Nation Under God.”

In February 2026, Hegseth invited Doug Wilson — a theologian who openly advocates for making the United States a Christian theocracy — to preach at the Pentagon.

DoD promotional videos during this period featured Bible verses overlaid on footage of fighter jets, tanks, and missiles.

In December 2025, Hegseth eliminated the Army Spiritual Fitness Guide because, in the department’s framing, it didn’t focus on God.

None of these actions changed a single line of the UCMJ. None rewrote an operational order. They changed the meaning environment in which orders are given and received. And they changed it at precisely the level where existing oversight instruments have no reach.

The mechanism Brian named

When Brian pointed to the New Apostolic Reformation, he was identifying something specific.

The NAR is a contemporary Christian movement organized around a doctrine called dominionism — the belief that Christians are called to take control of every sphere of society, including government and military institutions, as a precondition for Christ’s return. It emphasizes direct prophetic authority: leaders who claim to speak with divine mandate rather than institutional accountability.

Brian’s concern — and he said it plainly — runs deeper than bad theology. The NAR is an ideology that, by design, treats dissent as spiritual failure. “Demonizes those who differ” were his words. The language is structural. When authority is grounded in divine mandate, the cost of questioning it stops being professional and becomes existential.

That cost matters enormously when you’re asking whether a 22-year-old NCO will raise their hand in a briefing and say: “Sir, what is the legal basis for this mission?”

Research on sacred values helps explain why. Atran and Ginges, writing in Science, found that when values become sacralized — anchored to divine will or moral mandate — offering rational alternatives tends to increase resistance rather than reduce it. People don’t weigh sacred claims against secular ones. They treat the comparison itself as an outrage.

What this means practically: the standard repair move — pause, surface the competing meanings, pin a shared definition — does not work on a sacralized word. You cannot ask “what do you mean by order?” when the answer is “God’s command.” That question lands as a challenge to the frame itself.

Why repair can’t reach this from the inside

Most of the coordination failures I write about on this Substack follow a recoverable pattern. A word carrying multiple legitimate meanings collides with speed, pressure, and status differences. Nobody pauses to ask which meaning is active. The repair move is accessible — name the word, surface the split, pin a shared definition, log it for the next person who inherits the decision.

This failure mode is different in one specific way.

When a word’s meaning gets anchored to a sacred claim, all three repair activation conditions collapse simultaneously.

The first condition for repair is psychological safety — someone in the room has to believe that asking a clarifying question is safe enough to risk. In a military unit where questioning operational framing is coded as spiritual betrayal, that safety is gone before anyone opens their mouth.

The second condition is a functioning oversight channel. The UCMJ’s insubordination provisions, the whistleblower protection statute, the inspector general system — all of these still exist. They were designed for a meaning regime where order has a constitutional referent. They have no instruments for detecting that the referent has been replaced.

The third condition is the ability to surface competing definitions without triggering shutdown. Sacred values, by design, resist exactly that. The repair move that works in every other context becomes the very thing that escalates the conflict.

Brian could see all three of these conditions failing from the outside. He could call it clearly because he was looking at the theology as a theologian and the mechanism as someone who has spent decades watching what happens when sacred authority colonizes institutional authority.

From inside the institution, the same analysis requires naming a risk that the institution itself — at the level of its current leadership — has spent a year constructing the conditions to prevent anyone from naming.

What gets inherited if we don’t name it now

In a standard meaning drift scenario, the cost is rework, confusion, broken trust between colleagues. Recoverable, if painful.

When meaning drift reaches the word at the center of a system authorized to use lethal force — a system where the legal obligation to obey is written into statute, where questioning authority carries criminal jeopardy, where 22-year-olds are receiving operational briefings framed as prophecy — the cost scale is different.

Brian said “ridiculously dangerous” and meant it as precise diagnosis.

The institutional channels that should catch this — JAG officers, inspector generals, military chaplains, the whistleblower provisions in 10 U.S.C. § 1034 — are the subject of Part Two. So is the question Brian’s clarity actually raises.

If a minister on Facebook could see the mechanism clearly in a single scroll, what would it take for the institution to see it from the inside?

That is where this gets harder. That is where we are going next.

Part Two publishes soon. If this piece was useful to you, share it with someone who works in high-stakes environments — military, healthcare, aviation, emergency services. This is what meaning repair looks like when the stakes are real.

A note: Brian Cobb is quoted with his knowledge and permission. He holds a degree in biblical studies and has served in pastoral ministry for decades. You can find his work at Out of the Box Pastor.

One in three patients experiences a diagnosis-related communication failure. In business, independent assessments of identical insurance cases vary by 55%. The pattern is the same everywhere: meaning drifts, and nobody notices until the cost arrives.

Before you read, think of a decision your team made this week where you later discovered someone understood the outcome differently than you did. What happened?

The meeting where everyone agreed

Tuesday afternoon. Seven people around a table—or seven tiles on a screen. The project lead says, “We need to finalize the rollout plan by Friday.” Everyone nods. A few people type notes. The meeting ends on time, which feels like a win.

By Thursday, two things become clear. The engineering lead understood “finalize” to mean lock the technical spec. The marketing director understood it to mean confirm the launch date. The product manager thought “rollout plan” referred to the internal pilot; the sales lead thought it meant the customer-facing release. Nobody was confused during the meeting. Everyone left with a clear picture. The pictures were different.

This is what meaning failure looks like from the inside: total clarity that happens to be pointed in four directions at once.

What this course is about

This course teaches a set of small, repeatable moves for catching and repairing that kind of failure before it compounds. The moves come from decades of research across operating rooms, cockpits, military operations, and organizational teams. They work. The evidence is unusually clear on this point.

Here is the statistic that frames everything we’ll do together over the next ten weeks:

When directed, closed-loop communication is used in operating room emergencies, task completion reaches 100%. A sender states the request. The receiver repeats it back. The sender confirms. Every task gets done.

That practice appears in 12% of interactions.

The gap between 100% and 12% is the entire problem this course addresses. Teams already have the tools. The tools already work. What’s missing is the cognitive and social infrastructure that makes people actually use them—consistently, under pressure, when it matters most.

That infrastructure is what we’re building here.

The four phases of meaning failure

The MRCI framework maps how meaning breaks down and how it gets repaired. Four phases. Every communication failure your team has ever experienced fits somewhere in this sequence.

Phase 1: Drift. Meaning diverges silently. Assumptions go unchecked. A key term means one thing to you and something else to the person across the table. Context that was clear in one conversation doesn’t travel to the next one. The conditions for mutual understanding aren’t met, and nobody notices—because drift doesn’t announce itself. It feels like agreement. That Tuesday meeting? Pure drift.

Phase 2: Suppression. Here is where things get worse. Someone on the team senses the drift. A flicker of doubt: Wait, are we talking about the same thing? They consider saying something. They don’t. Maybe the meeting is moving fast. Maybe the person who would need correcting is senior. Maybe they’ve raised concerns before and nothing changed, so why bother. The doubt gets swallowed, and the drift continues unchecked.

Suppression is where most communication programs lose the game. They teach people what to say. They don’t address the forces that keep people from saying it.

Phase 3: Repair Activation. Someone speaks the first sentence. “Can I check something—when we say finalize, do we mean the spec is locked, or the date is confirmed?” That sentence is a repair move. It breaks the silence, surfaces the drift, and creates a moment where meaning can be realigned.

This transition—from suppression to activation—is the single point this course targets most aggressively, because it is where proven practices most consistently fail to get used.

Phase 4: Outcome. Meaning is restored. Or it isn’t. Either way, the result feeds back into the system. If the repair works—the team realigns, the ambiguity gets pinned, the meeting produces a shared picture—it becomes slightly easier for the next person to speak up next time. If the repair fails—the concern gets dismissed, the speaker gets a sideways look, the meeting steamrolls forward—it becomes harder. Failed repairs teach teams to suppress. Successful repairs teach teams to repair.

The cycle runs continuously. It is running in your team right now.

A note about this course’s own language

The terms I just introduced (drift, suppression, repair activation, outcome) are themselves subject to the process they describe. You read those words and built your own interpretation of what each one means. That interpretation was shaped by your experience, your professional context, and the conversations about communication you’ve had before this one.

For example: later in this course, I’ll talk about “shared definitions.” If you read that phrase and imagine I mean everyone in the room must agree on one permanent, universal definition of a key term before any work can begin — that’s drift happening right now, between us, on a concept about drift. What I actually mean is narrower and faster: before a decision is made, the people in the room check whether they’re using the same keyword to point at the same thing. That check takes ten seconds. The habit takes months.

I’m flagging this early because the pattern is predictive. Every term in the Meaning Repair Lexicon will land differently depending on what you bring to it. That’s normal. It’s also the exact phenomenon this course exists to address. When a term in this course feels wrong to you, that friction is data; it means your working definition, and mine haven’t been grounded yet. Treat that friction as an invitation to check, not a reason to dismiss.

Why drift is invisible: the grounding problem

In 1991, Herbert Clark and Susan Brennan published a theory of communication that explains why meaning fails the way it does. Their insight was deceptively simple: understanding doesn’t happen automatically when words leave someone’s mouth. It has to be built, move by move, between the people in the conversation.

Clark and Brennan described three requirements for what they called “grounding”—reaching mutual understanding:

Presentation. One person says something. States a plan. Makes a request. This is the part we’re all good at.

Acceptance. The other person provides evidence that they understood. A nod. A paraphrase. A question. A repeat-back. This is the part we skip.

Grounding criterion. Both people reach a mutual belief that understanding is sufficient for the current purpose. This is the part we assume happened when it didn’t.

Closed-loop communication—the practice with the 100% completion rate—formalizes the acceptance phase. The sender states the message. The receiver repeats it back in their own words. The sender confirms or corrects. Three steps. When all three happen, tasks get completed, and meaning holds.

When teams skip the acceptance phase (which is most of the time), they’re operating on an assumption of shared understanding that may or may not be accurate. In a low-stakes conversation, the cost of a wrong assumption is small. In an operating room, a cockpit, an emergency department, a product launch, a board decision, a custody negotiation—the cost can be enormous.

The WHO Surgical Safety Checklist, which formalizes a version of this grounding process, reduces mortality by 25%. Global compliance sits at 73%. The most meaning-critical phase of the checklist—the Sign Out, where the team confirms what was done—drops to 62%.

The tools exist. They work. The challenge is making them stick.

The 12% question

Why do teams suppress the use of practices they know are effective? The evidence points to a few reinforcing forces, and none of them involve laziness or ignorance.

Speed compresses the acceptance phase. Under time pressure, the grounding criterion gets lowered. “Good enough” understanding replaces verified understanding. This is rational in the moment and expensive over time.

Hierarchy suppresses repair activation. A 2024 study of 730 nurses found that power distance orientation significantly predicted whether nurses perceived speaking up as futile. When the person who needs correcting outranks the person who spotted the error, the calculus changes. The drift continues.

Silence is socially learned. A three-year longitudinal study at a German automotive plant found that organizational silence spreads through storytelling. People learn to stay quiet by hearing what happened to others who spoke up—even if those stories are years old, even if the manager in the story has long since left. Silence becomes cultural infrastructure, inherited through narrative.

Repeated futility produces acquiescent silence. When speaking up fails to produce change again and again, people stop trying. Two experiments with 654 participants demonstrated this pattern: voice futility produces a silence that mirrors learned helplessness. People don’t just choose not to speak. They lose the expectation that speaking could matter.

These forces operate simultaneously. They explain the 12%. They also explain why telling teams to “communicate better” doesn’t move the number.

The boundary this course is honest about

There is a condition that sits before drift, and it’s worth naming early.

Sometimes a person enters a conversation with a definition so settled that no verification attempt will register. They aren’t suppressing a repair move; they’ve decided repair isn’t needed. The word means what it means. The question was closed before it was asked.

I call this meaning refusal, and it’s the boundary of what meaning repair can reach.

Meaning refusal can look like prejudice, certainty, expertise, or exhaustion. It can come from a lifetime of being ignored when you raised concerns, until asking “what do you mean?” sounds like one more institution demanding you justify yourself. It can come from a position of genuine authority where you’ve been right so often that checking feels unnecessary. It can come from a fixed frame about a loaded term — “AI,” “equity,” “leadership,” “accountability” — where the definition was locked in years ago, and new information gets filtered through it rather than allowed to change it.

This course works in the space of engaged disagreement — where people differ but are willing to check. Recognizing when you’ve hit meaning refusal is itself a skill, and it saves you from spending repair energy where it can’t produce a result. We’ll return to this distinction in Module 4, when we look at suppression and silence.

For now, hold the question: when you imagine the person on your team who is hardest to reach with a clarifying question, is the barrier fear (suppression) or certainty (refusal)? The repair move for each is different.

What we’re building

Over the next ten weeks, you’ll build a set of tools—a Meaning Repair Operating System—drawn from your own workflows and conversations. Each module introduces one named move. Each move is small enough to use in a live meeting, a handoff, a Slack thread, or a family dinner.

The course follows the four phases. Modules 1 and 2 address Drift: how to spot it, name it, and diagnose where your team is most vulnerable. Modules 3 through 7 focus on Repair Activation: the specific sentences, protocols, and meeting moves that break silence and restore shared meaning. Modules 4 and 9 tackle Suppression and Outcome: the leadership practices and after-action reviews that determine whether repair becomes a habit or a one-time event. Module 10 builds the adoption plan that carries these moves past the end of the course.

You don’t need to learn something entirely new. You need to build the conditions for consistent use of what already works.

That’s a more tractable problem than it sounds. And it starts with a diagnostic.

Next post: The Meaning Risk Snapshot → A step-by-step assessment of where meaning is failing on your team right now, with a quantitative baseline you’ll re-measure at Day 30 and Day 60.

This is Module 1, Post 1 of Meaning Repair for High-Stakes Teams, a 10-module course on the What Time Binds Substack. If you haven’t already, start with the Meaning Repair Lexicon—your reference card for the language of meaning repair.

Three days ago, I published an essay called “The Inverse Problem,” about what it means when a for-profit company becomes the last guardrail against autonomous killing machines and mass surveillance. I argued that the roles of regulator and regulated had reversed — that a corporation was defending democratic values while the government demanded the right to override them.

I should note that I use Claude — the AI at the center of this story — alongside Google’s Gemini and NotebookLM in my daily work. I’ll address that directly at the end. For now, what matters is what happened in the seventy-two hours after that essay went live, because the story accelerated beyond anything I could have predicted, proving the thesis in ways that border on the absurd.

Friday: The Ban

You know the broad strokes from my first essay. Anthropic refused to remove two restrictions from its military contract: no fully autonomous weapons, no mass domestic surveillance. The Pentagon demanded access for “all lawful purposes.” Anthropic said no. Defense Secretary Hegseth designated the company a “supply chain risk” — a designation previously reserved for foreign adversaries like Huawei. Trump ordered every federal agency to cease using Anthropic’s technology.

What I did not know when I published is what the Pentagon had actually been asking for behind the scenes. According to Axios, even as Hegseth was tweeting the supply chain designation, a senior defense official was on the phone offering Anthropic a last-minute deal. That deal would have required allowing the collection and AI-processed analysis of Americans’ personal data — geolocation, web browsing history, financial information purchased from data brokers. Real surveillance, using specific, commercially available data about American citizens, processed at scale by artificial intelligence.

Anthropic said no to that, too.

Friday Night: The Replacement

Hours after the ban, OpenAI CEO Sam Altman announced that his company had reached its own deal with the Pentagon for classified deployment. Altman said the agreement included the same red lines Anthropic had demanded: no mass surveillance, no autonomous weapons.

Here is the detail that matters more than anything else in this story, and I want you to read it carefully.

Anthropic’s contract demanded absolute prohibitions. The restrictions existed regardless of what any other law said. OpenAI’s contract permits the Pentagon to use its technology for “all lawful purposes” — and its red lines only apply where existing law already forbids the action.

This distinction is the entire game.

Collecting commercially available personal data on American citizens — your location, your browsing habits, your financial records — is already legal. The government has been buying this data from brokers for years. Lethal autonomous weapons systems are not firmly prohibited by any U.S. statute; they are only partially constrained by a Defense Department directive that the Defense Department itself can revise at any time. As one analysis of the contract language put it: the wording “repeatedly qualifies OpenAI’s prohibitions as dependent on existing restrictions. Anything not forbidden is permitted.”

So when OpenAI says “no mass surveillance” but qualifies it with “consistent with applicable law,” and the applicable law already permits mass data collection, what exactly has been prohibited? When OpenAI says “no autonomous weapons” but conditions it on regulations the Pentagon writes for itself, what exactly has been restricted?

Altman admitted on social media that the deal was “definitely rushed” and that “the optics don’t look good.” He was right about the optics. Whether he was right about the substance depends entirely on contract language that is not fully public and may never be.

After the contract details were published, Techdirt’s Mike Masnick argued the deal “absolutely does allow for domestic surveillance,” because it references compliance with Executive Order 12333 — a Reagan-era directive that intelligence agencies have long used to collect communications data on Americans by tapping lines outside the country.

The Inverse Problem just gained a new dimension: the roles of corporation and government have reversed, and now the language of safety itself can be deployed as a simulacrum — an image of restraint that functions as its opposite.

Saturday: The War

On February 28, the United States and Israel launched Operation Epic Fury against Iran. Strikes hit nuclear facilities, military infrastructure, and a government compound in Tehran. Ayatollah Ali Khamenei was killed.

And according to the Wall Street Journal, Axios, and Reuters, U.S. Central Command used Claude throughout the operation — for intelligence assessment, target identification, and simulation of battle scenarios. The same tool the president had banned the day before. The same company he had called “left-wing nut jobs.” The same technology the Pentagon had designated a supply chain risk equivalent to a foreign adversary.

Defense officials told reporters that removing Claude from the operational workflow was simply not possible on short notice. The model is embedded in classified systems in ways that cannot be unwound in a day, or a week, or even six months without significant risk to ongoing operations.

Consider what this means. On Friday, the U.S. government declared that Anthropic’s technology was so dangerous to national security that every contractor in the country must sever ties with it. On Saturday, the U.S. government used that same technology to prosecute the most consequential military operation of the year. One of these things is a lie. Both of them are true.

The R Street Institute’s Mark Dalton identified the core contradiction: the Pentagon considered Anthropic’s technology so vital to national defense that it threatened to invoke the Defense Production Act to retain access — and then, days later, designated that same company a supply chain risk. Dalton warned that the next time this designation is applied to a company with actual ties to a foreign adversary, its credibility will be diminished.

He is right, and he understates the problem. The credibility is gone entirely. The supply chain risk designation has been revealed as what it always was in this context: a political weapon dressed up as a security assessment.

Saturday and Sunday: The People Voted

While the government was banning Anthropic and using its technology to fight a war, something remarkable happened in the consumer market. Claude hit number one on Apple’s U.S. App Store, overtaking ChatGPT for the first time. A “Cancel ChatGPT” movement spread across Reddit and X, with users posting guides for deleting their accounts and migrating to Claude. Anthropic reported that free users increased over 60% since January, with daily sign-ups quadrupling.

The American public did in forty-eight hours what Congress has failed to do in a decade: it rendered a verdict on the ethics of AI deployment. With downloads. With app deletions. With the only ballot available.

This is both encouraging and damning. Encouraging because it suggests that when the stakes are made clear, people will act on their values. Damning because the fact that consumer behavior is the only functioning feedback mechanism in the most consequential technology debate of our time is itself a symptom of institutional failure.

Congress still has not held a hearing. No legislation has been introduced. The only democratic accountability in this entire saga has come from people tapping an icon on their phones.

Friday Through Sunday: The Employees

There is one more thread, and it may be the most revealing.

Before the deadline passed on Friday, more than 300 Google employees and over 60 OpenAI employees signed an open letter urging their companies to stand with Anthropic. The letter warned that the Pentagon was trying to divide the industry: “They’re trying to divide each company with fear that the other will give in. That strategy only works if none of us know where the others stand.”

Google DeepMind’s Chief Scientist, Jeff Dean, publicly endorsed the letter’s position, writing that mass surveillance violates the Fourth Amendment. More than 100 DeepMind employees sent a separate internal letter to leadership demanding restrictions on military applications.

And then Sam Altman signed the Pentagon deal anyway.

The fracture is stunning. OpenAI employees signed a letter saying the Pentagon’s demands were unacceptable. Hours later, their CEO accepted those demands. The Inverse Problem has moved inside the walls of the corporation itself — between the people who build the technology and the people who decide what to do with it.

What Seventy-Two Hours Proved

When I wrote the first essay, the Inverse Problem was a framework — a way of understanding a structural shift in which the expected roles of power and restraint had reversed. Three days later, it has become an empirical observation, confirmed by events:

A government banned a technology it cannot stop using. A company walked away from hundreds of millions of dollars rather than compromise on two principles almost every American would agree with. A competitor claimed identical principles while accepting contract language that may render those principles meaningless. Workers at that competitor’s company publicly opposed what their leadership did in their name. And the public, lacking any institutional mechanism for input, voted with the only ballot available to them — an app download.

The Inverse Problem is a condition. And seventy-two hours proved that the condition is deepening.

The Question That Now Has a Price Tag

On the same day Anthropic was blacklisted and OpenAI took the Pentagon deal, OpenAI announced a $110 billion funding round — $50 billion from Amazon, $30 billion from Nvidia, $30 billion from SoftBank — at a $730 billion pre-money valuation.

I am noting a coincidence that the market will inevitably interpret. One company said no to power and lost its government business. The other said yes and secured the largest private funding round in technology history, on the same day, from the companies that will build the physical infrastructure of AI for the next decade.

The Inverse Problem now has a price tag. Anthropic’s two red lines — a human in the loop before anyone dies, and no mass AI surveillance of American citizens — cost the company $200 million in government contracts and potentially its entire federal business ecosystem. OpenAI’s willingness to accept “all lawful purposes” language coincided with the infusion of more capital than most nations’ GDP.

The market is watching. Every AI company in the world is watching. And what they are learning is this: safety has a cost, and compliance has a reward, and the distance between them is measured in hundreds of billions of dollars.

What the Mirror Shows Now

I said in the first essay that the Inverse Problem is a mirror. What I see reflected now, seventy-two hours later, is worse than what I saw before.

I see a government that will use the word “risk” as a weapon against a company for exercising the same judgment the government relied on the next morning. I see a deal dressed in the language of safety that may contain no safety at all. I see workers at one of the world’s most powerful companies publicly opposing a decision their leadership made without them. I see a public that cares deeply and has almost no institutional channel through which to express it. And I see a technology — one that helped plan airstrikes on Tehran — that is too important for anyone to control and too dangerous for no one to try.

The Inverse Problem has become as real as it gets. It has a body count, a price tag, and an app store ranking.

And it is only Monday.

What Time Binds is a column about the moments where past assumptions collide with present realities. If you read the first essay, share this one. The people who need to understand the Inverse Problem are the ones who haven’t heard of it yet.

Full disclosure: I use Claude, Google’s Gemini, and NotebookLM in my daily work and am a paying subscriber to both Anthropic’s and Google’s AI products. I have no financial relationship with either company beyond those subscriptions. This essay was written with Claude’s assistance.

This module includes an interactive course orientation. Tap the image above to explore the full MRCI framework, the 4-phase model, and the 10-module course map before you begin.

Learning Objective

Learners diagnose their team’s current communication vulnerabilities using a structured, evidence-based assessment and establish a quantitative baseline they will re-measure at Day 30 and Day 60.

Recall Prompt (Prior Knowledge Activation)

Before you read, think of a decision your team made this week where you later discovered someone understood the outcome differently than you did. What happened?

Substack Post Sequence

Post 1: Why Meaning Fails Silently (Concept + Why)

1,200–1,500 words | ~10 min read

Introduces the MRCI 4-phase model (Drift → Suppression → Repair Activation → Outcome). Uses the 100%-vs.-12% closed-loop statistic as the course’s motivational anchor: teams don’t need new tools, they need consistent use of proven ones. Explains meaning drift using Clark & Brennan’s grounding theory. Frames the course’s purpose: building cognitive infrastructure for consistent use of what already works.

Post 2: The Meaning Risk Snapshot (How + Practice)

1,000–1,400 words | ~8 min read

Step-by-step instructions for completing the Meaning Risk Snapshot. Includes 5–10 adapted TAT items (Ali et al., 2024; Cronbach’s α = 0.97) scored on a Likert scale for quantitative baseline. Adds a Cultural Context field using Meyer’s eight dimensions as a lightweight self-assessment (power distance, communication directness, comfort with silence). Practice rep: complete the snapshot for your primary team.

Post 3: Repair Rep + Discussion

600–800 words | ~5 min read

Sets the habit formation expectation: “This is a 10-week build. You will not master these moves by Module 3. You will feel awkward. That’s the signal that you’re building a new habit, not that you’re failing.” Transfer Bridge prompt: Where outside your current work role would this diagnostic have changed an outcome this week? Discussion thread opens.

Deliverable

D1: Meaning Risk Snapshot — Completed team assessment with TAT-adapted quantitative items, cultural context field, and narrative risk summary. Scored baseline to be repeated at Day 30 and Day 60.

Example Part 2 Cultural Context Map

68KB ∙ PDF file

Download

Download

D1 Meaning Risk Snapshot Fillable

46.8KB ∙ PDF file

Download

Download

Deliverable Reflection Prompts

• What strategy did I use to complete this deliverable?

• What was most difficult and why?

• What would I do differently next time?

Validated Instruments Referenced

Team Assessment Tool (Ali et al., 2024), Meyer’s Culture Map dimensions, ROCI subscales (Handley et al., 2024)

Eight AM, Saturday, February 28, 2026—The waffle was covered in whipped cream and strawberries. I was sitting in a diner in Ontario, California, reading David Frum’s Atlantic piece on my phone, and the strawberries were that bright diner red that doesn’t occur in nature. The TV above the counter had the sound off. Closed captions scrolled: Tehran. Explosions. Operation Epic Fury. A woman two booths over was laughing about something on her own phone. A kid was kicking their legs back and forth, dangling off a chair.

I kept reading. I kept eating. And I kept feeling something I’ve felt before and have only recently learned to name: the sensation of being in two places at once. Here, in the territory—a diner, a Saturday, a waffle getting cold. And there, in the map—the abstraction called “war,” now being projected onto a country of ninety million people by a president who posted a video on Truth Social telling Iranian civilians that “the hour of your freedom is at hand.”

I deployed to Iraq twice in 2003 and again in 2008. Afghanistan in 2011 and 2013. Five combat tours as a Marine combat engineer. I built camps and tore them down in both countries. I watched Saddam's statue come down, and then I watched what came after—the looting, the power vacuums, the slow unraveling that no briefing slide had anticipated. By the time I came back at the end of 2003, the place looked nothing like the victory map. When I came back in 2008, we were still building things in the wreckage of what we'd promised. I know what "freedom" sounds like in an operational plan. I know what it looks like on the ground eighteen months later, when the power grid is down, and the interpreters who trusted you are trying to get their families out.

So when I sat in that diner reading Frum’s article—an intelligent piece, a well-constructed piece—I felt both attached and detached. Attached because I know what it costs when a country decides to do this. I’ve carried that cost in my body. Detached because the essay I was reading, for all its sharpness, was still operating on a map. The territory was somewhere else entirely.

Subscribe now

The Map and the Territory of “Opportunity”

Frum’s article runs on a clean binary: opportunity (free Iran) vs. danger (unchecked Trump). He structures the whole thing around that split. The frame is doing a lot of work, maybe too much.

Here’s where I agree. The Iranian people’s courage is beyond dispute. Millions rose against their government. Tens of thousands died. The Human Rights Activists News Agency estimates over 7,000 killed. The International Centre for Human Rights puts the number at 43,000. Security forces flooded residential neighborhoods on motorbikes, showing off their guns to create fear. Hospitals were overwhelmed. The regime imposed the longest internet blackout on record to keep the world from seeing what it was doing. Those who could communicate with the outside world asked for help. That is real. Those are real people.

Here’s where I start asking questions. Frum writes as if the “opportunity” becomes self-evident once the regime’s military assets are degraded. As if the word “freedom” can travel from a Truth Social video to the streets of Tehran on the back of a cruise missile and arrive meaning what we think it means.

This is a Ghost Map.

I use that term in my book to describe abstractions that have detached from the messy territory they once tried to represent. “Freedom,” as deployed in Trump’s video statement, functions as a permission structure wearing the clothes of a destination. I’ve seen this before. I’ve stood inside it.

In 2003, the map said “liberation.” The territory said something else. Looted ministries. Collapsing infrastructure. Sectarian fault lines that the briefing slides never mentioned. An insurgency that metastasized in the gap between what we promised and what we built. I watched the statue come down. I was there for every month that followed, as the gap between the word “liberation” and the reality on the ground widened into something that swallowed four thousand American lives and two trillion dollars. Rep. Lloyd Doggett, in his statement today, said what I’ve been saying for twenty years: the statue falling was the beginning of chaos, the easy part. The hard part was everything that came after—and nobody had a plan for it.

Iran is almost four times larger than Iraq. Its military is far more capable. Its regional alliances run deeper, and its asymmetric capabilities make escalation unpredictable even from limited strikes. And no one—not Trump, not Frum, not the Pentagon—has described what the day after looks like.

Frum acknowledges the uncertainty—“the outcome is obviously extremely uncertain”—and then keeps writing as if the upside is real, and the downside is manageable. That’s the structure of adolescent risk assessment. I’ve seen it in operational planning sessions. The people who’ve never been downrange draw clean arrows on the map. The people who have been downrange stare at the blank spaces between the arrows and ask: what happens there?

What Does “Without Congress” Actually Mean?

Here’s where Frum’s analysis gets sharper, and where my framework meets his argument squarely.

Trump launched this war without congressional authorization. The Constitution assigns the power to declare war to Congress. The War Powers Resolution of 1973 requires the president to notify Congress within 48 hours and bars deployment beyond 60 days without legislative approval. What happened instead: Secretary of State Rubio called the Gang of Eight shortly before the strikes began. The Armed Services Committees were notified after the bombs were already in the air.

This pattern is decades old. Obama intervened in Libya for eight months without asking Congress. Both Bushes used AUMFs—Authorizations for the Use of Military Force—as open-ended permission slips. Trump, in his first term, cited the 2002 Iraq AUMF to justify killing an Iranian general. The last time Congress actually declared war was 1942. Against Bulgaria.

So when Frum writes that Trump started this war “without even a figment of congressional authorization,” he’s right—and the honest version of that sentence has a longer tail: within a system that has been allowing presidents to start wars without real congressional authorization for the better part of a century.

This is the Korzybskian Gap in its institutional form. Our tools for making war scale at the speed of a cruise missile. Our tools for deciding whether to go to war still run on a document written when muskets took thirty seconds to reload. The Constitution anticipated deliberation. The national security state was built for speed. Every president—Democrat and Republican—has chosen speed.

I think about this gap every time I hear someone say “Commander in Chief” as if that title were self-authorizing. The title describes a role inside a system of constraints. The Constitution says the president commands the military that Congress authorizes—a relationship that has drifted so far from its original design that most Americans, and most members of Congress, have forgotten what the document actually requires.

The bipartisan war powers resolution from Reps. Khanna and Massie had been introduced before the strikes. It was already going to be a hard vote. Now, with bombs falling and service members at risk, it becomes nearly impossible. The machine works this way by design. You launch the war before the vote, and then the vote becomes a referendum on whether you “support the troops.” I’ve been one of those troops. Five times. I can tell you that “supporting” us means asking the hard questions before we get on the plane—asking them while there’s still time to change the answer.

The rally-around-the-flag effect—first described by political scientist John Mueller in 1970—is a known, studied, predictable phenomenon. Leaders experience short-term spikes in approval at the onset of international crises. Matthew Baum’s research at Harvard found that the spike is driven largely by independents and opposition-party members shifting toward the president. The more divided the country, the larger the rally effect.

So here’s the engineering question: Who benefits from launching a war eight days after the Supreme Court struck down your tariffs 6-3 and called them unconstitutional? Who benefits from starting a regional conflict five days after delivering a State of the Union speech that, as Frum notes, was “speckled with insults and stunts”?

The cause can be legitimate and the timing can be strategic and the institutional structure can be broken. Adult analysis holds all three. Adolescent analysis picks one and yells.

The Zoological Error in Real Time

Frum writes a line that stopped me mid-bite: “No president in American history has shown himself less trustworthy with power than Donald Trump.”

Strong claim. And I think it’s doing something Frum doesn’t fully intend. It personalizes a structural problem.

In my book, I call this the Zoological Error—the mistake of treating political systems as if they’re dominance hierarchies in a captive wolf pack, where all that matters is who the alpha is and whether the alpha is benign or malicious. The alpha framing makes great copy. It makes terrible institutional design.

The deeper question: Why does the system hand any single person the unilateral ability to launch a war that could kill hundreds of thousands of people, restructure the Middle East, and drain trillions of dollars—all before Congress reconvenes from recess?

Five deployments taught me something about leadership and systems. The best units I served in depended on structures—pre-mission checklists, after-action reviews, standing operating procedures, clear escalation protocols—that made good decisions likely regardless of who was in charge. The worst units ran on personality. When the CO was sharp, things worked. When the CO was reckless, people got hurt. A coin flip with lives on the table.

When Frum writes that “the ideal palliative would be for House Speaker Mike Johnson and Senate Majority Leader John Thune to act like the independent constitutional officers they are supposed to be,” he’s describing a repair that depends entirely on individual character. That’s the Gardener’s nightmare. You don’t build a bridge and then hope the bolts feel like holding. You design the structure so it holds whether the bolts are heroic or mediocre.

Rep. Warren Davidson, a Republican, said it today: “I have asked for a classified briefing defining the mission in Iran. In the absence of new information, I will support the War Powers resolution.” Rep. Thomas Massie, also Republican, called the strikes “acts of war unauthorized by Congress.” These are institutional immune responses—weak ones, from individual actors, pointing toward what an adult system would require: structural checks that don’t depend on courage, character, or the next election cycle.

The Paradox at the Center

Here’s the question that Frum’s article circles without ever landing on: How do you build systems powerful enough to handle the mess we’re in without handing some maniac the keys?

Frum wants the Iranian regime gone. He also wants the American president constrained. He recognizes that the man delivering “freedom” to Iran is the same man who, in Frum’s own words, tried to send members of Congress to prison for making a video reminding members of their duty to obey the law. That tension runs through every paragraph of his essay, and he never resolves it—because he can’t, within the frame he’s built.

What Frum leaves out matters more than what he puts in. His piece, for all its intelligence, has no structural proposal. No institutional repair. He calls for “a dozen Republican House members—even two or three senators—who remember the principles they supposedly believe.” A plea for individual virtue inside a system that has spent decades selecting against that virtue.

This is where the Engineer-Gardener framework matters. The Engineer in me sees the War Powers Resolution and says: this tool has been broken since the day it was passed. Every president has ignored it. No enforcement mechanism. Congress has discovered that passivity is electorally safer than assertion. The tool fails. Redesign it.

The Gardener in me sees the cultural layer underneath. A nation trained—by media cycles, by rally-around-the-flag dynamics, by decades of executive overreach from both parties—to treat war as the president’s prerogative and congressional oversight as an inconvenience. A statute alone won’t repair that. The repair lives in rebuilding the civic infrastructure that makes citizens capable of demanding something better—in classrooms, in newsrooms, in the cognitive immune system of a population running on Ghost Maps about what “strength” looks like.

The Repair

So what would adult institutional design look like in this moment?

First, the War Powers Resolution needs teeth. Real teeth. An automatic funding cutoff if Congress doesn’t affirmatively authorize military action within a defined window. No more “we’ll vote on it next week” while the 60-day clock runs and the war becomes a fait accompli. Senator Tim Kaine has been pushing versions of this for years. The structure should make inaction equivalent to a “no” vote—silence should mean stop, the same way it works in a pre-mission brief when someone doesn’t confirm ready.

Second, the “Gang of Eight” briefing model—where a handful of leaders get a call from the Secretary of State hours before bombs drop—is notification dressed up as oversight. Adult oversight means standing committees with real-time access to operational planning before the decision is made. Senator Mark Warner said it plainly: Trump’s characterization of the situation during the State of the Union didn’t match Warner’s understanding from classified briefings. That gap between the public narrative and the classified reality is exactly the kind of meaning divergence that my research on meaning repair identifies as a precursor to catastrophic institutional failure.

Third, and this is the harder one: we need to stop treating the rally-around-the-flag effect as a natural law and start treating it as a design vulnerability. The effect is real and short-lived—typically thirty days to two months, per Lian and Oneal’s 1993 findings. It is driven by the temporary absence of elite criticism and the surge of in-group identification. In a healthy system, that window closes quickly as oversight kicks in, media scrutiny deepens, and citizens process information. In a degraded system—one running on noise, bandwidth taxes, and a media environment optimized for engagement over accuracy—the window stays open long enough for the war to become irreversible.

The Korzybskian Gap will not close by itself. That’s the core argument of my research. Our tools for making war got faster. Our tools for deciding whether to go to war stayed slow. The brains tasked with making those decisions are running on fumes—taxed by noise, stressed by scarcity, herded into partisan cages that make nuanced thought feel like betrayal. Just read my article from The Inverse Problem: When a For-Profit Company Becomes America’s Last Guardrail Against Killer Robots.

What I Mean

I paid for the waffle. I left the diner. The TV was still running silent footage of smoke over Tehran.

Outside, the parking lot was full of people doing Saturday things. Getting gas. Loading groceries. Checking phones. None of them were on a map that included “major combat operations in Iran.” They were in the territory—the ordinary, ongoing territory of a country that is, at this moment, starting a war most of its citizens didn’t vote on, weren’t consulted about, and may not fully understand for weeks.

Frum sees the paradox. A brutal regime being struck by an untrustworthy president. An opportunity wrapped in a threat. He sees it clearly enough to name both poles.

What he misses—what almost no one in mainstream commentary catches—is the machine. The machine that allows one person to launch a war. The machine that converts rally effects into political cover. The machine that selects for leaders who exploit the gap between the speed of weapons and the speed of deliberation. The machine that treats the question “should we go to war” as a matter of presidential temperament rather than institutional architecture.

David Frum wrote an intelligent article about a dangerous man with too much power. I’m writing about the system that keeps handing the keys to whoever wins the cage match.

The Iranian people deserve freedom. American service members deserve a government that follows its own Constitution before putting their lives at risk. And all of us—every person whose life is shaped by decisions made at the speed of a cruise missile—deserve institutions designed for adults.

We’re not there yet. We’re adolescents with god-level tools and animal-level governance.

The question is whether we grow up before the next war, or because of it.

I’ve already lived through five tours of “because of it.” I’d like to try the other way.

Jerry W. Washington, Ed.D., is a Marine combat engineer veteran, independent researcher, and author of the forthcoming The Adulthood of Humanity (2026). He writes What Time Binds on Substack.

If this framework changed how you see the news, share it. If it made you uncomfortable, that’s the biology working correctly.

Something broke in the American experiment this week, and almost nobody has the framework to understand it. So let me give you one.

In political science, there’s a concept you could call the Inverse Problem — when the actors you expect to play one role swap places entirely with the actors you expect to play the other. Regulators become deregulators. The regulated become the regulators. The machine designed to protect you becomes the thing you need protection from, and the machine designed to extract profit from you becomes the thing standing between you and catastrophe.

That is exactly what happened on February 27, 2026, when the President of the United States threatened a private AI company with “the full power of the presidency” — including criminal consequences — because it refused to build autonomous killing machines and mass surveillance tools for the Pentagon.

The company is Anthropic, maker of the Claude AI system. And here is the detail that should stop you cold: Anthropic is a for-profit company.

Subscribe now

The Facts, Because They Matter

Let’s establish what actually happened before we talk about what it means.

Anthropic signed a contract worth up to $200 million with the Pentagon last summer. Claude became the first AI model to operate on the military’s classified networks, deployed through a partnership with Palantir Technologies. By all accounts, the arrangement was working. Claude was helping warfighters with logistics, planning, analysis — the kinds of applications that make modern defense function.

Then, in January, reports emerged that Claude had been used during the U.S. military operation that captured Venezuelan president Nicolás Maduro in Caracas. Anthropic couldn’t confirm or deny the specifics — classified operations are classified — but the company raised concerns that Claude’s use may have crossed lines that its terms of service explicitly prohibit.

That triggered a chain reaction. Defense Secretary Pete Hegseth summoned Anthropic CEO Dario Amodei to a meeting. According to multiple reports, military officials warned they could invoke the Defense Production Act — a Cold War-era law — to seize broader authority over the technology. Hegseth gave Anthropic until 5 PM Friday to remove all restrictions on military use of Claude.

Anthropic’s response came in a public statement: they cannot permit Claude to be used for two specific purposes. Just two. Fully autonomous weapons — weapons that kill without a human being in the decision chain. And mass domestic surveillance — the wholesale collection and AI-processed analysis of Americans’ personal data.

That’s it. That was the line. Use our AI for anything else. Plan operations. Analyze intelligence. Optimize logistics. Build better systems. Just don’t build robots that kill people on their own, and don’t spy on the entire American population with it.

The Pentagon’s response? Hegseth designated Anthropic a “supply chain risk.” President Trump announced a six-month phaseout of all Anthropic products across the federal government and posted on Truth Social, calling the company “left-wing nut jobs” who had made “a disastrous mistake trying to strong arm the Department of War.”

The Pentagon’s Undersecretary for Research and Engineering, Emil Michael, called Amodei “a liar” with “a God-complex” who “wants nothing more than to try to personally control the US Military.”

For saying no to killer robots and mass surveillance.

The Inverse Problem

Here is where you need to pay attention, because this is the moment the world turned upside down.

For the past decade, the dominant anxiety about artificial intelligence has gone something like this: powerful technology companies would build increasingly dangerous AI systems, and governments — slow, bureaucratic, and ultimately accountable to voters — would need to step in and regulate them. The corporation would push for maximum capability. The government would push for maximum safety. That is the normal polarity of the relationship between power and restraint in a democracy.

What happened this week is the precise inversion of that dynamic.

A for-profit corporation — one that stood to make $200 million from this single contract alone — voluntarily limited its own product. It drew lines around what its technology should not do. No law required it. No regulator ordered it. The people who built the technology simply understood, better than anyone in government, what it was capable of becoming.

And the United States government, the entity constitutionally charged with providing for the common defense while securing the blessings of liberty, demanded that those safety limits be removed. When the company refused, the government moved to destroy it commercially.

Read that again if you need to.

The corporation said: We believe a human being should be in the chain of command before anyone is killed.

The government said: Remove that restriction, or we will end you.

Why “For-Profit” Is the Detail That Changes Everything

Anthropic competes in the open market. It raises venture capital. It answers to investors. Specifically, Anthropic is a Public Benefit Corporation — a for-profit entity that has raised billions from Google and Salesforce, investors who expect returns.

Its corporate structure includes something called a Long-Term Benefit Trust, an independent body that can select and remove board members to ensure the company stays aligned with its mission. It was deliberately designed so that safety considerations couldn’t be easily overridden by profit motives. And still — Anthropic has revenue targets, employees with stock options, and competitors like OpenAI, Google, and Meta breathing down its neck.

And it just walked away from a $200 million government contract — and potentially its entire federal business — because it wouldn’t agree to let its technology kill people without human oversight.

If Anthropic were a nonprofit, this would be admirable. The fact that it is a for-profit company makes it extraordinary. A company that exists to generate shareholder value decided that some values cannot be sold. That there is a price too high, even when you’re in the business of being paid.

When was the last time you saw that happen in America?

The Silence of Congress

Ben Rhodes, who served as deputy national security advisor under President Obama, put it plainly in a recent interview: in any normal era, this is exactly the kind of issue that would produce legislation. Congress would hold hearings. International negotiations would establish norms. Treaties would be drafted — the way we drafted the Chemical Weapons Convention, the way we built the Nuclear Non-Proliferation Treaty.

Instead, Congress has done nothing. The international community is fractured. And the only regulation that exists on the most powerful technology humanity has ever built is the terms of service written by the companies that built it.

Think about that for a moment. The only thing standing between the United States military and fully autonomous AI weapons systems is a corporate policy document (of one company). No law. No treaty. No constitutional provision. A terms of service agreement from a company in San Francisco.

And the government just tried to rip it up.

The Domino That Didn’t Fall — Yet

Here is the part of the story that offers a thin, complicated kind of hope.

Within hours of Trump’s announcement, OpenAI CEO Sam Altman sent a memo to his employees declaring that OpenAI shares Anthropic’s red lines. “We have long believed that AI should not be used for mass surveillance or autonomous lethal weapons,” Altman wrote, “and that humans should remain in the loop for high-stakes automated decisions.”

This is significant. OpenAI and Anthropic are fierce competitors. Altman and Amodei have deep personal and professional tensions that trace back to Amodei’s departure from OpenAI in 2021. For Altman to publicly align with Anthropic on this issue amounts to an entire industry drawing a line in the sand.

The solidarity is also fragile. The Pentagon has reportedly already approved OpenAI’s proposed safety red lines in a new arrangement. The government’s position may already be softening — or the government is simply shopping for a more compliant partner. The dynamics are shifting by the hour.

What the Inverse Problem Tells Us About Where We Are

The Inverse Problem extends well beyond this particular news cycle. It is a structural feature of the moment we are living through.

When the institutions designed to protect the public interest instead pursue unchecked power, and the institutions designed to pursue profit instead protect the public interest, you are not living in a system that is functioning correctly. You are living in a system where the wiring has been reversed.

I want to be precise about what I’m saying here: neither actor is playing the role it was designed for. We need democratic governments that regulate powerful technology in the public interest. We need corporations that innovate within boundaries set by democratic accountability. What we have instead is a government demanding the right to build autonomous killing machines and a corporation begging it to stop.

Experts at Lawfare note that using the Defense Production Act to force Anthropic to strip its safety guardrails would be “without precedent.” The DPA allows the government to prioritize its contracts — to jump to the front of the line. It has never been used to force a company to change the fundamental nature of its product. Forcing Anthropic to remove its ethical restrictions would raise First Amendment questions: the government would be compelling a company to express values it rejects.

Anthropic has said it will challenge the supply chain risk designation in court, calling it “legally unsound” and warning it would “set a dangerous precedent for any American company that negotiates with the government.”

We are in genuinely uncharted territory, both technologically and constitutionally.

The Question This Moment Is Asking You

Here is what I keep coming back to.

The people who built Claude — the engineers, the researchers, the safety teams at Anthropic — know more about what this technology can do than anyone in the Pentagon. They know more about it than Pete Hegseth. They know more about it than Donald Trump. They have spent years studying the failure modes, the edge cases, the scenarios where AI systems behave in ways their creators did not intend.

And those people are scared enough that they built restrictions into their own product, at the cost of their own revenue, knowing it might cost them everything.

When the people who understand the technology best are the ones most afraid of what it can do, and the people who understand it least are the ones most eager to unleash it, that should tell you something. That should tell you everything.

Anthropic’s two red lines are not radical positions. A human should decide before a machine kills someone. The government should not use AI to conduct mass surveillance on its own citizens. These are ideas that, stated plainly, almost every American would agree with. They are the bare minimum of civilized restraint.

And yet here we are, watching a for-profit company fight the federal government to preserve them.

The inverse problem is a mirror. And what it’s reflecting back at us right now is a country where the last line of defense for democratic values might be a terms of service agreement, written by a company that technically exists to make money.

If that doesn’t keep you up at night, you’re not paying attention.

What Time Binds is a column about the moments where past assumptions collide with present realities. If this piece resonated, share it. The inverse problem only gets worse when people aren’t watching.