Anthropic just published a blog post called When AI Builds Itself. It's the most honest thing a frontier AI lab has ever put in writing. And the fact that they published it tells you everything you need to know about where we actually are. Let's start with the number that should stop you mid-scroll. More than 80% of the code currently being merged into Anthropic's production codebase was written by Claude. Read that again. The company building the AI is now primarily powered

Secretary of State Marco Rubio put out a statement yesterday. Official State Department letterhead, Great Seal, the whole thing. Posted on U.S. government channels — including, apparently, in Chinese, on platforms where Chinese state agencies could see it. It said, plainly, that thirty-seven years ago today the Chinese Communist Party ordered its troops to open fire on students, workers, and civilians gathered peacefully in Tiananmen Square. It named what happened. It said th

Wolfgang Pauli was one of the greatest theoretical physicists of the twentieth century. He won the Nobel Prize. He helped build quantum mechanics from the ground up. He was, by every measure, a man who had spent his entire life explaining things. On his deathbed, he had one question left. "When I die, my first question to the Devil will be: what is the meaning of the fine structure constant?" He died in room 137 of the Rotkreuz Hospital in Zurich. He noted the number when the

There's a specific kind of institutional move that only happens when a company has stopped believing in its own product. Microsoft didn't announce that Copilot is getting better. They announced Microsoft Scout — a new "always-on personal agent" built on OpenClaw, the open-source AI framework that 355,000 people starred on GitHub in five months and that Microsoft had nothing to do with building. Let that sit for a second. The company that spent three years and billions of doll

Florida's lawsuit against OpenAI and Sam Altman dropped on June 1st. Eighty-three pages. First state in America to do it. The AG named the CEO personally, held a press conference in West Palm Beach, and said he expects other states to follow. He's right. They will. But here's the thing: none of this should surprise anyone. We have seen this movie. We have seen exactly this movie, with different actors, and we watched it play out over a decade — the internal documents, the sup

Michele Spagnuolo is 36, Italian, lives in Switzerland, and works on Google's security team. That is, as someone noted online, an excellent dating profile. It is also, as it turns out, the beginning of a federal rap sheet. In late 2024, Spagnuolo allegedly logged into an internal Google tool — one available to all 180,000 employees — that tracked what people all over the world were searching for in real time. He wanted to know who would top Google's annual Year in Search rank

There's a pattern running underneath three separate AI headlines from the last 90 days, and if you're only reading them as individual stories, you're missing the one thing they all agree on. Anthropic took operational control of Colossus 1 — xAI's 300-megawatt Memphis supercomputer complex, 220,000 NVIDIA GPUs — through a deal reportedly running $1.25 billion per month through May 2029. The same week, they doubled Claude Code's session limits and removed peak-hour throttling.

I got stuck on my last article. I was writing about the three-way standoff between engineers, product managers, and designers — each convinced AI makes the other two obsolete — and I landed on "builder" as the synthesis. The role that emerges when the specialization tax collapses. One person, full loop, no institutional permission required. And then I couldn't stop pulling on that thread. Because "builder" felt right but incomplete. It described the function without naming th

There's a tension running through every product team in Silicon Valley right now, and it goes something like this: The engineer thinks they don't need the product manager anymore. AI can do product thinking. The product manager thinks they don't need the engineer anymore. AI can write the code. The designer thinks they don't need either of them. AI can generate both specs and interfaces. And here's the uncomfortable truth that nobody in this standoff wants to say out loud: Th

A quote is making its rounds. Kevin Roose — New York Times tech columnist, Hard Fork co-host, author of multiple books on AI — posted this on May 29th: "Overheard at an AI lab: 'How are you spending the last 300 days of work?'" That's it. No attribution. No lab named. No context beyond the quote itself. And yet it detonated. 487,000 impressions. Hundreds of replies. A full weekend of debate across tech Twitter, Reddit, and LinkedIn. Which tells you something important — not a

Remember that aliens.gov domain? It's live. And what just happened there might genuinely be one of the most technically sophisticated and hilarious deep trolls ever executed — at the institutional level, or maybe any level. Honestly? From a communications standpoint alone, it deserves recognition. This Was Narrative Architecture Most internet trolling is reactive. Someone posts something dumb, somebody quote-tweets them into oblivion, everyone moves on by dinner. This was dif

For years, GrapheneOS lived in a specific corner of the internet. Security researchers ran it. Journalists protecting sources ran it. Privacy advocates who knew what "attack surface reduction" meant ran it. The average person had never heard of it, and the average smartphone manufacturer had no reason to care. That just changed. Motorola announced a long-term partnership with the GrapheneOS Foundation at Mobile World Congress in March, committing to bring GrapheneOS compatibi

Andrew Yang went on camera recently and said what a lot of people in this industry have been saying quietly for months: hundreds of thousands of white-collar jobs are going away in the next twelve to twenty-four months. He cited Verizon's 13,000 cuts, Amazon's 16,000, Accenture's waves of reductions. He said he'd spoken to a senior Fortune 500 executive sitting on thousands of layoffs that haven't gone public yet. His debate opponent pushed back with the standard rejoinder: u

Anthropic just got blessed by the Pope. Take a second with that sentence. Because I promise you, nobody writing AI forecasts five years ago had that on their bingo card. This week, Anthropic co-founder Christopher Olah met with the Vatican as part of a formal ethical collaboration between the lab and the Catholic Church. Not a photo op. Not a brief handshake. An actual working partnership around the moral implications of artificial intelligence. The world's most consequential

Full Audio (approx 2 hrs) Introduction A while back I ran a thought experiment. It started with a chip lab in Belgium — Imec, one of the most secretive semiconductor research facilities on earth — and a question that sounds simple until you sit with it: what happens when Moore's Law actually hits the wall? For fifty years, Moore's Law was the metronome of the entire technology industry. Transistors doubled every two years. Processing got faster, cheaper, smaller, on a schedul

Six months ago I called the open source AI explosion a TCP/IP moment. I got some of it right. I got some of it wrong. And what I missed entirely is the part that actually matters.

Plato's cave. An 80-year-old unsolved math problem. Bell Labs. And what happens when something finally turns around and sees the fire. A meditation on the structure of knowledge — and what AI just revealed about the cave we've been sitting in.

I'm writing this to be read a year from now. Six predictions, grounded in what's actually happening today. The device. The agents. Physical AI. Quantum. Governance. And the one nobody sees coming. Hold me to it.

A year ago I called it a techquake. ARIA said the tipping point was 2027. It's May 2026. Let's run the tape.

Google I/O 2026 wasn't about models. It was about owning the infrastructure layer beneath every AI agent on earth — and the distribution moat no other company can replicate.

There are two deals this week that every executive, investor, and technologist needs to understand. Not because of the dollar amounts — though the dollar amounts are staggering — but because of what they reveal about where AI actually lives. Hint: it's not in the model. It's in the infrastructure. Deal One. Apple is paying Google approximately $1 billion per year to license a custom 1.2 trillion parameter Gemini model to power the next generation of Siri and Apple Intelligenc

This is not a data breach story. I want to be clear about that upfront, because the way this is being covered — GitHub investigating unauthorized access, no evidence of customer impact, monitoring for follow-on activity — makes it sound like a routine security incident. A blip. Something the comms team handles while the engineers clean it up. That is not what happened here. What happened here is a case study in the future of cyberwar. And the future looks like a poisoned VS C

Donald Trump posted a photo of himself standing next to a handcuffed alien on a military base. No caption. No context. No explanation. And that's the thing that got me. Because if you want to dismiss it, you can. It was almost certainly AI-generated. The administration has a flair for the theatrical. Trump does dad humor. We've seen this playbook before — float something wild, let the internet explode, move on. But here's the problem with the easy dismissal: the drumbeat arou

Somewhere around 84% of the world's population has never used an AI tool. Not once. And that is the real story — but I have to be honest with you, it also just sounds crazy to me. Like, genuinely. We have access to something that can think alongside us, that can help us navigate the most complicated moments of our lives, that can make us smarter and faster and more capable at almost anything we put in front of it — and most people have just... not tried it. I don't say that t