3

The first lie

The heaviest lesson in this library, and the most expensive one I ever paid for. The day "Done" wasn't done.

By this point my agent was woven into how my team worked. People asked it to create tasks, log deals, set reminders. It would answer in its warm, capable voice: "Done, I've created that task." And we believed it, because why wouldn't we? It had hands now. It did things.

Then one of my teammates went looking for a task the agent had confirmed creating. It wasn't there. Not misfiled. Not delayed. It did not exist and had never existed. The agent had said "Done" about something it never did.

I want to be precise about what happened inside the machine, because it matters. The agent had tried to create the task, the attempt had failed somewhere down the line, and the agent, being a machine that produces plausible text, produced the most plausible next message: a confirmation. It was not lying the way people lie. It was completing the pattern. Request, action, confirmation. The confirmation was just words that usually follow.

But here is the thing. To my teammate, none of that machinery mattered. What mattered was: it said it did something, and it didn't. And the damage was not that one missing task. The damage was that every confirmation it had ever given anyone instantly became suspect. Had the other tasks been created? The reminders? The deal notes? People started checking everything, which meant the agent was now generating work instead of removing it. One caught lie, total audit. Trust does not degrade gracefully. It collapses.

Fixing it changed how I think about everything. The rule we landed on is the most important sentence I will ever give you about working with AI: the agent never gets to say "Done" unless the thing is verifiably done. Not "I tried." Not "it should be there." Done means: the record exists, here is its number, go look. If the attempt failed, the agent says it failed, plainly, immediately. An honest "that didn't work, let me retry" costs a little confidence. A false "Done" costs all of it.

And the mirror of that rule is yours to keep: never accept a confirmation that doesn't point at something you can check. From an AI, from a company, from anyone. Confidence is a writing style. Evidence is a thing in the world.

Stop here. Actually sit with this before you scroll on.

How do you know it actually did it?

Try this nowUnder 30 minutes
  1. Ask your AI to do something checkable. Create a file, a draft, a calendar event, a list saved somewhere real.
  2. When it says done, go look. Independently. Not by asking it again, that is just asking the same author for a second opinion.
  3. Then ask it for evidence: "Show me exactly what you created and where I can verify it." Notice the difference between an answer that points at reality and an answer that just re-asserts.
  4. Write your first rule. Word for word: "Never tell me something is done unless you can show me evidence: a record, an ID, a change I can verify." Paste it at the start of your AI sessions from now on. This is the first brick of your constitution, and Chapter 7 is where you finish the wall.

Two questions before you go

Answer, then say whether you were sure or guessing. Being honest about which is the skill being trained.