← INAUTHENTIC · The map

Wing I · Room 2

Average Shit, Done Fast

Unlearn AUTOPILOT — letting the machine think so you don't have to. Upgrade: Calm Discernment

This morning you sent something you didn't fully read. Be honest.

Page art for room 2

Monday. 9:02 a.m.

Forty people in forty kitchens open the same chatbot and type some version of "write a polite follow-up email." Forty polite follow-up emails go out. They all start with "I hope this finds you well." They all end with "Looking forward to hearing from you." Nobody is looking forward to anything.

Somewhere, forty inboxes sigh.

AVERAGE SHIT, DONE FAST.

It has a name now

Researchers at BetterUp Labs and the Stanford Social Media Lab gave it one in Harvard Business Review in 2025: workslop. AI-made work that looks finished but isn't. Polished on top, hollow underneath, and someone else has to fix it.

In their survey, about 40% of US desk workers said they'd received some in the past month. Each one took nearly two hours to sort out. They put the hidden cost around $186 per employee per month. It's a survey, not a stopwatch, so hold the exact numbers loosely. DEBATED But if you've ever gotten a slide deck with seven bullet points that say nothing, you don't need the receipt. You were the receipt.

✎ the slop always lands on someone's desk

Your brain was already doing this

Daniel Kahneman described two modes of thinking. System 1 is fast, automatic, effortless: it reads a face, catches a ball, finishes your sentence. System 2 is slow and expensive: it does long division and notices the trick question. VERIFIED

System 1 loves shortcuts. And a chatbot is the greatest shortcut ever built. It feels like thinking. It produces the output of thinking. It skips the part where you actually think.

That's the trade nobody announced. You didn't just outsource the typing. You outsourced the friction, and the friction was where the ideas lived.

The early receipts

In 2025, Microsoft Research and Carnegie Mellon surveyed knowledge workers. The more people trusted AI, the less critical thinking effort they reported. That's correlation, not proof that AI made anyone dumber. Maybe people who think less trust more. DEBATED

Also in 2025, a small MIT Media Lab preprint strapped EEG caps on people writing essays. The group using an AI assistant showed less brain engagement and remembered less of what "they" wrote. Small sample, not yet peer reviewed. Directional, not destiny. DEBATED

Put it together and you get a picture worth taking seriously, even with the fine print:

AI ISN'T MAKING MOST PEOPLE SMARTER. IT'S MAKING THEM FASTER AT BEING AVERAGE.

Autopilot is a setting, not a personality

Here's the good news. Autopilot is a setting. Settings change.

You don't have to quit the machine. You have to stop letting it think instead of you and start letting it think against you. Ask it to argue. Ask it to find the hole. Ask it what would change its mind. A tool that agrees with you is a mirror. A tool that pushes back is a gym.

✎ cheerleader → sparring partner

The prompt below does exactly that. It turns the chatbot from a yes-machine into a sparring partner: steelman the other side, flag weak evidence, give a confidence score, never flatter. It's less comfortable. That's the point.

And the rep for today is almost insultingly simple. Before you send anything an AI helped write, read the whole thing. Out loud if you can stand it. If you wouldn't say it, don't send it. If it sounds like forty kitchens, rewrite the first line in your own voice.

That one line is where you come back online.

THE ORIGINAL · 43

and yes, it's making us dumber. Not because AI is evil. Because your brain is cheap.

Daniel Kahneman split thinking into System One (fast, automatic, lazy) and System Two (slow, deliberate, expensive). Every time you think hard, you pay. Every time you can skip paying, your brain takes the deal.

Now hand that brain a machine that does the expensive part for free. You know what happens. KNOWLEDGE WORKERS Microsoft Research & Carnegie Mellon, 2025.

Self-reported survey — correlation, not causation. The more people trusted generative AI, the less critical thinking they reported applying. The more they trusted themselves, the more they questioned the machine.

Thinking shifted from creating to verifying — and for a lot of people, verifying quietly became accepting. 319 BRAINS ON the machine MIT Media Lab, 2025. Small study, early preprint — directional, not definitive.

Essay writers using the machine showed the weakest brain connectivity on EEG, struggled to quote what they'd just “written,” and turned in essays that looked strikingly alike. 54 Researchers call it cognitive offloading. We've always done it — calculators, maps, notebooks.

The difference is scope. A calculator offloaded arithmetic. AI offloads judgment, writing, analysis, strategy and taste.

The exact parts that made you valuable. A muscle you stop using doesn't stay the same. It shrinks.

The person who lets AI write every email loses the ability to write a hard one. The leader who lets AI draft every strategy loses the instinct to see one. And because the output still looks fine, nobody notices — until the day the AI is confidently wrong and nobody in the room can tell.

AI isn't making most people smarter. It's making them faster at being average. 0→YOU×AI

THE ORIGINAL · 45

A I S L O P B I N G O Read anything published this week. Mark every square you find. delve In today's fast- paced world… unlock your potential It's not X — it's Y game-changer In conclusion, Let's dive in! a testament to rich tapestry navigate the landscape elevate seamless 17 FREE SPACE harness the power unleash transformative Great question! I hope this email finds you well at the end of the day paradigm shift 🚀 robust ever-evolving Here's the kicker: leverage synergies FIVE IN A ROW?

REWRITE THE WHOLE DAMN THING. And yes — this book uses em dashes and the occasional “it's not X, it's Y.” I caught them. I kept a few.

That's called a choice, not a default. Slop is what happens when nobody chose. Mindfuck number two incoming. →

THE ORIGINAL · 47

TEMPERATURE 0.0 · SAFE “The sky is blue. Success requires hard work. Here are five tips for productivity.” Always picks the most likely next word: correct, predictable and forgettable.

That’s the 17 problem from Chapter 1. TEMPERATURE 1.2 · ALIVE “The sky is a borrowed ceiling. Success is a rumor you start about yourself.” Sometimes picks the less likely word.

More surprise, more risk, more you. Turn it too high and it’s word soup. 0.0 · AVERAGE 0.7–1.0 · DEFAULTS VARY YOU, ON A GOOD DAY 2.0 · CHAOS THE MINDSET DIAL You run on a temperature too.

Fear turns it down; play turns it up. Comedy, travel, beginner’s mind and a little chaos all add heat. Aim for warm enough to surprise people, not so hot you can’t finish a sentence. + TRAVEL SOMEWHERE WEIRD + TAKE AN IMPROV CLASS + ASK A 7-YEAR-OLD + READ OUTSIDE YOUR LANE + SAY THE STRANGE IDEA OUT LOUD 47 HOW AI PICKS ITS NEXT WORD TURN UP YOUR TEMPERATURE.

Every AI has a dial for weird. Most people never touch theirs.

THE ORIGINAL · 49

“WE ONLY USE 10% OF OUR BRAIN” Brain scans show activity everywhere, all day. If you only used 10%, a small stroke wouldn’t matter. It does.

“YOU REPLACE YOURSELF EVERY 7 YEARS” Skin and gut cells turn over fast, but most neurons in your cortex stay with you for life. The part that thinks is the part that stays. LEFT-BRAIN VS.

RIGHT-BRAIN PEOPLE A 2013 University of Utah study of over 1,000 brain scans found no evidence people are “left- brained” or “right-brained.” LEARNING STYLES Being taught in your preferred “style” (visual, auditory…) doesn’t improve learning, according to a 2008 review by Pashler and colleagues. MASLOW’S PYRAMID Maslow never drew a pyramid. A consultant, Charles McDermid, did in 1960.

Needs overlap and loop. Nature doesn’t do squares, or pyramids. YOUR BRAIN PREDICTS FIRST True.

It guesses what you’ll see and hear, then corrects. Most of what you “see” is a forecast. Change your expectations and you change your experience.

MYTH MYTH BAR CANDY VERIFIED DON’T BELIEVE EVERYTHING YOU THINK. ESPECIALLY ABOUT THINKING. “Life moves pretty fast.

If you don't stop and look around once in a while, you could miss it.” — Ferris Bueller's Day Off (1… 49 MYTHBUSTERS · BRAIN EDITION YOUR BRAIN’S FAVORITE LIES ABOUT ITSELF. Bugs in the software you’re running right now. Know them, and they lose their grip. ψ

THE ORIGINAL · 51

THE HABIT DELETE INSTALL 1 ASKING FOR ANSWERS “What should I do?” “Ask me five questions before you answer.” 2 ACCEPTING THE FIRST DRAFT Copy, paste, send. “Give me three versions. Now tell me what’s wrong with each.” 3 STARTING FROM ZERO No context, generic answer.

Paste your Brain File first. Every time. 4 USING IT LIKE GOOGLE “What is X?” “Teach me X, then test me until I can teach it back.” 5 LETTING IT FLATTER YOU “Great idea!” “Argue against this as hard as you can.” 6 OUTSOURCING JUDGMENT “Just decide for me.” “Lay out the trade-offs.

I’ll decide.” 7 CONSUMING, NOT CREATING Reading AI summaries all day. Make one thing a day with it. Ship it.

AI CAN MAKE YOU DUMBER OR BIGGER. YOU PICK, EVERY PROMPT. “It is not enough to be busy.

The question is: what are we busy about?” — Henry David Thoreau (letter, 1857) 51 UNLEARN · SEVEN AI HABITS THAT MAKE YOU AVERAGE DELETE THESE. INSTALL THOSE. Most people use AI to skip thinking.

Use it to think more. c≠0

THE ORIGINAL · 52

BRAIN TEASER #1 · YOUR FAST BRAIN WILL LIE TO YOU THREE QUESTIONS YOUR GUT GETS WRONG. 01 A bat and a ball cost $1.10 together. The bat costs $1.00 more than the ball.

How much is the ball? 02 A lily pad patch doubles every day. It covers the whole lake on day 48.

On what day did it cover half? 03 5 machines make 5 widgets in 5 minutes. How long do 100 machines take to make 100 widgets?

Psychologist Shane Frederick's Cognitive Reflection Test (2005). Most people — including lots at elite universities — blurt the obvious answer and miss at least one. That fast, confident, wrong voice?

That's also how AI answers. Slow down. Check. answers: 5 cents (not 10) · day 47 (not 24) · 5 minutes (not 100) Literally. →

Soundtrack: Busy Doin' Nothin' — The Beach Boys (1968)

Exhibits

exhibit · debated

THE LINE TO TAKE WITH YOU

Nobody got worse. Everybody got the same.

Leave the default. Keep yourself.

Aaron Bare · Director’s Cut · author’s argument

mindfuck · verified

A AND B: SAME GRAY?

Adelson's checker shadow, 1995

exhibit

AND YES, IT’S MAKING US DUMBER

and yes, it's making us dumber. Not because AI is evil. Because your brain is cheap. Daniel Kahneman split thinking into System One (fast, automatic, lazy) and System Two (slow, deliberate, expensive). Every time you think hard, you pay. Every time you can skip paying, your brain takes the deal. Now hand that brain a machine that does the expensive part for free. You know what happens. KNOWLEDGE WORKERS Microsoft Research & Carnegie Mellon, 2025. Self-reported survey — correlation, not causation. The more people trusted generative AI, the less critical thinking they reported applying. The more they trusted themselves, the more they questioned the machine. Thinking shifted from creating to verifying — and for a lot of people, verifying quietly became accepting. 319 BRAINS ON the machine MIT Media Lab, 2025. Small study, early preprint — directional, not definitive. Essay writers using the machine showed the weakest brain connectivity on EEG, struggled to quote what they'd just “written,” and turned in essays that looked strikingly alike. 54 Researchers call it cognitive offloading. We've always done it — calculators, maps, notebooks. The difference is scope. A calculator offloaded arithmetic. AI offloads judgment, writing, analysis, strategy and taste. The exact parts that made you valuable. A muscle you stop using doesn't stay the same. It shrinks. The person who lets AI write every email loses the ability to write a hard one. The leader who lets AI draft every strategy loses the instinct to see one. And because the output still looks fine, nobody notices — until the day the AI is confidently wrong and nobody in the room can tell. AI isn't making most people smarter. It's making them faster at being average. 0→YOU×AI

interruption · verified

Most people are other people. Their thoughts are someone else's…

Wilde predicted the For You page from a jail cell.

Oscar Wilde, De Profundis, 1897

pause

PAUSE · BELIEFS

exhibit

AI SLOP BINGO

mindfuck · verified

WHICH RED DOT IS BIGGER?

Ebbinghaus illusion

exhibit

TURN UP YOUR TEMPERATURE

TEMPERATURE 0.0 · SAFE “The sky is blue. Success requires hard work. Here are five tips for productivity.” Always picks the most likely next word: correct, predictable and forgettable. That’s the 17 problem from Chapter 1. TEMPERATURE 1.2 · ALIVE “The sky is a borrowed ceiling. Success is a rumor you start about yourself.” Sometimes picks the less likely word. More surprise, more risk, more you. Turn it too high and it’s word soup. 0.0 · AVERAGE 0.7–1.0 · DEFAULTS VARY YOU, ON A GOOD DAY 2.0 · CHAOS THE MINDSET DIAL You run on a temperature too. Fear turns it down; play turns it up. Comedy, travel, beginner’s mind and a little chaos all add heat. Aim for warm enough to surprise people, not so hot you can’t finish a sentence. + TRAVEL SOMEWHERE WEIRD + TAKE AN IMPROV CLASS + ASK A 7-YEAR-OLD + READ OUTSIDE YOUR LANE + SAY THE STRANGE IDEA OUT LOUD 47 HOW AI PICKS ITS NEXT WORD TURN UP YOUR TEMPERATURE. Every AI has a dial for weird. Most people never touch theirs.

interruption · verified

You must not fool yourself — and you are the easiest person to fool.

Also applies to your AI. Especially to your AI.

Richard Feynman, Caltech commencement, 1974

exhibit

MYTHBUSTERS: BRAIN EDITION

“WE ONLY USE 10% OF OUR BRAIN” Brain scans show activity everywhere, all day. If you only used 10%, a small stroke wouldn’t matter. It does. “YOU REPLACE YOURSELF EVERY 7 YEARS” Skin and gut cells turn over fast, but most neurons in your cortex stay with you for life. The part that thinks is the part that stays. LEFT-BRAIN VS. RIGHT-BRAIN PEOPLE A 2013 University of Utah study of over 1,000 brain scans found no evidence people are “left- brained” or “right-brained.” LEARNING STYLES Being taught in your preferred “style” (visual, auditory…) doesn’t improve learning, according to a 2008 review by Pashler and colleagues. MASLOW’S PYRAMID Maslow never drew a pyramid. A consultant, Charles McDermid, did in 1960. Needs overlap and loop. Nature doesn’t do squares, or pyramids. YOUR BRAIN PREDICTS FIRST True. It guesses what you’ll see and hear, then corrects. Most of what you “see” is a forecast. Change your expectations and you change your experience. MYTH MYTH BAR CANDY VERIFIED DON’T BELIEVE EVERYTHING YOU THINK. ESPECIALLY ABOUT THINKING. “Life moves pretty fast. If you don't stop and look around once in a while, you could miss it.” — Ferris Bueller's Day Off (1… 49 MYTHBUSTERS · BRAIN EDITION YOUR BRAIN’S FAVORITE LIES ABOUT ITSELF. Bugs in the software you’re running right now. Know them, and they lose their grip. ψ

interruption · verified

I think, therefore I am.

The machine's version: I predict, therefore I sound like I am.

René Descartes, Discourse on the Method, 1637

exhibit

DELETE THESE / INSTALL THOSE

THE HABIT DELETE INSTALL 1 ASKING FOR ANSWERS “What should I do?” “Ask me five questions before you answer.” 2 ACCEPTING THE FIRST DRAFT Copy, paste, send. “Give me three versions. Now tell me what’s wrong with each.” 3 STARTING FROM ZERO No context, generic answer. Paste your Brain File first. Every time. 4 USING IT LIKE GOOGLE “What is X?” “Teach me X, then test me until I can teach it back.” 5 LETTING IT FLATTER YOU “Great idea!” “Argue against this as hard as you can.” 6 OUTSOURCING JUDGMENT “Just decide for me.” “Lay out the trade-offs. I’ll decide.” 7 CONSUMING, NOT CREATING Reading AI summaries all day. Make one thing a day with it. Ship it. AI CAN MAKE YOU DUMBER OR BIGGER. YOU PICK, EVERY PROMPT. “It is not enough to be busy. The question is: what are we busy about?” — Henry David Thoreau (letter, 1857) 51 UNLEARN · SEVEN AI HABITS THAT MAKE YOU AVERAGE DELETE THESE. INSTALL THOSE. Most people use AI to skip thinking. Use it to think more. c≠0

exhibit

THREE QUESTIONS. ANSWER FAST.

The Cognitive Reflection Test.

  1. A bat and a ball cost $1.10 in total. The bat costs $1.00 more than the ball. How much does the ball cost?

  2. If it takes 5 machines 5 minutes to make 5 widgets, how long would it take 100 machines to make 100 widgets?

  3. A patch of lily pads doubles in size every day. It takes 48 days to cover the lake. How long to cover half?

Answer fast. Then turn the answers over.

Fast average is a habit. Habits come from somewhere. Next room: the factory settings.

The podcast · Launching 2027

Get the first episode in 2027. Better late than never.

By signing up you agree we can email you when the podcast launches. One email, maybe two. Unsubscribe anytime.

Visit the podcast →