The bottleneck has moved from writing code to securing it - and that shift is a business opportunity.
The Mythos-to-Fable hype cycle, the ban, and the compute economics that quietly make these models dumber.
Dawkins says he thinks Claude is conscious. Everyone called him a senile old man. I think he nailed it.
Where the title of Amodei’s famous AI essay comes from - a 1967 Richard Brautigan poem, read aloud.
ID and age checks are coming to AI, and OpenAI is turning into a kind of token bank you have to register with.
AI does not just automate tasks - it moves human work to whatever constraint is now scarce.
A live conversation with Giuseppe Funicello on whether writing code, learning frameworks, and building products still pays off.
Why the US government restricted Anthropic Fable and Mythos models, and why I think surveillance is the real story.
I'm on Italian public radio talking Anthropic, the SpaceX valuation, and what AI means for programmers.
Why premium AI models are sold "by the sip" - and why the "too dangerous" safety framing is really about B2B pricing.
Anthropic keeps calling for an AI slowdown - but the simpler explanation is they fell behind on compute.
A neuroscientist pushes back on mind-body dualism: intelligence is physical, machines understand, and we are beautiful biological machines.
The new encyclical on AI is sharper and more political than I expected - and it's a little embarrassing that the Pope has to be the one saying it.
Reading the new AI encyclical, building a Pope-vs-Big-Tech video game with Claude Code, and finding the exact limits of vibe coding.
