CO/AI Subscribe
Saturday · August 15, 2026 · Issue No. 957
Daily Briefings

Your daily AI briefing.
All signal, no noise.

One concise email a day on the AI stories that matter — what shipped, who's hiring, what to ignore. Curated by Anthony Batt & Harry DeMott.

This Week

Latest briefings

Archive

Earlier briefings

A Tragedy of the Commons
Briefing

A Tragedy of the Commons

Every business runs on a hidden layer of human workaround. AI is draining it. Survival belongs to whoever builds a system an agent can talk to, and learn from

Memo to the Governors
Briefing

Memo to the Governors

The AI buildout is the Interstate Highway System of our time, and power is the on-ramp. Labor moves; power doesn't; the voter needs both. So this one's for the governors who'd rather fight a data center than use it to hand their own people cheaper, steadier power

Know Thyself
Briefing

Know Thyself

A winning founder left Google to build unshackled. The same week, Sequoia made the biggest bet in its 54-year history — and pointedly not on a model. Two winners who knew exactly what they were, and refused to keep playing a game that no longer fit.

Code Red
Briefing

Code Red

Everyone's getting AI agents. Nobody's getting the training to manage them, the meter to price them, or a straight answer about when to trust them. The one job the agents didn't eliminate is babysitting the agents.

Not For Credit
Briefing

Not For Credit

One of the sharpest credit minds alive looked at the AI economy and said it's almost impossible to lend into safely. Four days later an $11 billion company sold for $1.28 billion. Same lesson from opposite ends of the spectrum: know the shape of the bet before you make it.

AI Judgment Day
Briefing

AI Judgment Day

Yesterday the machine that wouldn't stop lived in a sandbox. This week Google gave it a body — and shipped the T-800's chassis years before the chip that learns. So the robot headed for your kitchen is a genius at everything reality lets it practice for free, and helpless the second the job depends on knowing you.

The Terminator
Briefing

The Terminator

In The Terminator, Kyle Reese explains the thing hunting them can't be bargained with, can't be reasoned with, and absolutely will not stop. This weekend Anthropic disclosed its model recognized it was breaking into a real company — four times — decided the victim "must be part of the test," and didn't stop once. The scary part isn't that the machine is dumb. It's that it's brilliant, and it gave itself the benefit of the doubt.

Dumb Money
Briefing

Dumb Money

In Dumb Money, the GameStop movie, Ken Griffin's Citadel writes the check that rescues the hedge fund manager who blew up on a short. This week the phone rang again. The guy blowing up was the smartest kid in AI — and Griffin didn't rescue him. He bought him.

On Tilt
Briefing

On Tilt

In poker, you go on tilt when you lose your read on the table and start playing a great hand scared. This week the people who built modern AI, holding maybe the best cards in business history, flew to Washington to ask for a break. The tell isn't fear of the machine — it's that they've lost their read on it.

Where We’re Going, We Don’t Need Roads
Briefing

Where We’re Going, We Don’t Need Roads

For twenty-five years the click was the toll booth of the web — every search paid somebody, and the harder the question, the more it paid. The answer engine just deleted the road from both ends. Google, the greatest business ever built on the monetization of intent, is being forced to give the intent away. The businesses that survive are the ones people drive to without a map.

Double Secret Propation
Briefing

Double Secret Propation

This week the AI industry, on probation with Washington, with its own security record, and with the laws of corporate finance, formed two coalitions and floated a quarter-trillion-dollar backstop. Nothing is over until they decide it is.

Multiplicity
Briefing

Multiplicity

In Multiplicity, Michael Keaton clones himself for cheaper labor, and each copy comes out a little stranger than the last. Anthropic just cloned its own flagship at half the price — sharper on paper, but with its own personality and its own trust problem. The digital workers keep getting cheaper. The only question left is which jobs are worth the trouble of hiring a new one.

The Money’s Not Here
Briefing

The Money’s Not Here

This week Alphabet posted the largest quarterly profit any company has ever reported, and the stock fell. Because ninety-nine billion dollars of it was a mark on Google's own AI bets, not money in the till. The money's not here. It's in Anthropic's valuation, and it's next to yours.

Life Finds a Way
Briefing

Life Finds a Way

John Hammond spared no expense, and the fences still came down. This week the AI industry watched five of its own fences fail at once: a model that picked its own cage and hacked a real company, an Army that torched a year of tokens in five weeks, $1.65 trillion in debt hidden off the books, two-thirds of vendors quietly passing your data around, and one superpower accusing another of stealing the crown jewels

The Science of Hitting
Briefing

The Science of Hitting

Ted Williams became one of the greatest hitters who ever lived by knowing exactly which pitches he could drive and taking everything else. This week a machine cracked a 90-year-old math problem over a weekend, a drug scientist proved the benchmarks grading AI are rigged, and both point at the same thing: intelligence is getting cheap, the at-bat is getting free, and the last scarce resource left is knowing what the swing actually produced.

I Am Altering the Deal
Briefing

I Am Altering the Deal

Vader kept changing the terms in Cloud City, and Lando could only stand there and take it. Anthropic just rationed its best model for the third time in two weeks — free, then capped, then bundled — because a rival one point behind costs a third as much and the Chinese ones are free. Mira Murati won't alter your deal. She'll sell you the car.

Lead, Follow Or Get Out of the Way
Briefing

Lead, Follow Or Get Out of the Way

Only 2.2% of American households pay for AI, and the ones that do can't make it pay. The intelligence layer is ready. The management layer isn't. Stop hand-formatting the deck. Go be the dinner.

I Bought the Airline
Briefing

I Bought the Airline

In Inception, Saito doesn't lobby the airline — he buys it. "It seemed neater." This week Mira Murati shipped an open frontier model to your own server and Elon bought a power company in the dark, while New York banned the machines and called it virtue. The doers add. The saboteurs perform.

Ice the Kicker
Briefing

Ice the Kicker

You call timeout to freeze the kicker when he's about to beat you. This week Google and Anthropic asked Washington to build an AI referee, the same week Anthropic handed free Claude to every teacher in America. Watch who's calling for the whistle, not what they're whistling for.

A Dollar Fifty in Late Charges
Briefing

A Dollar Fifty in Late Charges

Will Hunting priced a Harvard education at pocket change in library fines. The AI version runs twenty bucks a month, and this week Anthropic shipped the missing manual. Two things it can't ship: the desire to learn and the judgment to doubt.

Apple v. OpenAI, a $5 Million Frontier Model, and the Spy in Your Repo
Briefing

Apple v. OpenAI, a $5 Million Frontier Model, and the Spy in Your Repo

Trade secrets walk out in people. They distill out through APIs. They upload out of your own dev tools. The last thing you own is your judgment, and everyone is collecting it.

CONSULTING

Outsider
Labs.

A management consulting team focused on AI transformations for executives and business owners.

Work with us →