REALITYWIPE

OLD SCRIPT DETECTED

Reaching for the AI proves I never learned it

Try this response:

My hand reaching fast isn't a verdict on my memory — it's what happens after the manual step stops being needed.

Take the 60-second self-check

SEE THE PRACTICE

Build a response for this thought—and one move to prove it.

THE THOUGHT

Reaching for the AI proves I never learned it

YOUR RECORDED RESPONSE

My hand reaching fast isn't a verdict on my memory — it's what happens after the manual step stops being needed.

ONE PRIVATE MOVE

Next time you reach for the AI, pause for one minute after reading its output and privately write down the one line you'd change and why.

The half-second before your hand hits the AI tab

What you do with the next reach

The decision isn't to stop reaching for the AI — it's to stop letting the reach itself be the verdict. Next time your hand moves, let it move, then catch the moment right after: read the output like a colleague's draft, not a rescue. Mark, even mentally, the one line you'd change and why. That's the part no reflex can fake, and it's the part worth noticing on purpose.

TRACE · 01/04

How this script runs you

  • You clock how fast your hand moved to the AI tab and treat the speed itself as proof you didn't know the answer.
  • You solve something correctly using AI output but still feel like the win doesn't count as yours.
  • You replay old projects looking for the moment you 'should have' learned this well enough to skip reaching at all.

SOURCE_LOCATED · 02/04

The hidden rule underneath

The AI Obsolescence Script

If knowledge were real, it would surface from memory before my hand moves — any reach for outside help is proof the learning never actually landed, only looked like it did.

FORGING_REPLACEMENT · 03/04

Replacement lines (record these)

  • My hand reaching fast isn't a verdict on my memory — it's what happens after the manual step stops being needed.
  • Judging the output in two seconds is the skill. The AI didn't generate that judgment.
  • I can let the reach happen and still ask what I'd change, on purpose, right after.

Generated per person in the app — these are the flavor, not your script. Yours is built from your exact words.

INSTALL · 04/04

The protocol, on one card

When I think

Reaching for the AI proves I never learned it

I say

My hand reaching fast isn't a verdict on my memory — it's what happens after the manual step stops being needed.

Then I do one thing

Next time you reach for the AI, pause for one minute after reading its output and privately write down the one line you'd change and why.

STATUS: READY_TO_INSTALL

Wipe this thought

4 minutes. Your voice. Free.

Start this exact wipe in the app →

The wipe protocol
  1. Write the thought exactly as it plays: "Reaching for the AI proves I never learned it". Word for word — the wipe targets the sentence, not the vibe.
  2. Trace the rule and the avoided action. What does this thought conveniently excuse you from doing?
  3. Record the replacement lines below in your own voice. Speak like you mean it — no recording, no install.
  4. Run the loop: play it every morning and night, log one proof action a day for 7 days.

Straight answers

If the reach toward the AI happens before I even think, doesn't that prove the knowledge isn't really mine?

Speed isn't a lie detector for learning. Fast reflexes usually mean a step got automated, not that it never existed — the same way an experienced driver's foot finds the brake before conscious thought does. What you do with what comes back is where your own knowledge shows up, and that part still requires you.

How is judging an AI's output in a few seconds different from just not having learned the material?

Judging output fast draws on the exact same base your slower, from-memory answers used to draw on — pattern recognition built from problems you've personally debugged before. Someone with no grounding in the material can't tell a plausible-looking wrong answer from a right one that quickly. The speed of your judgment is evidence of prior learning, not its absence.

What if I genuinely can't reconstruct the answer without looking at what the AI produced first?

That gap is worth noticing, but it's a gap in unaided recall under time pressure, not proof the underlying understanding is fake. Recall speed and comprehension are different capacities. You can name the one line you'd revise in an AI's answer, which requires understanding recall alone can't test.

Related old scripts

Related self-checks

Related science

Researchers to explore

Related mechanisms

The research behind this script