OpenAI, long a regulation skeptic, now asks the US for binding AI safety rules — the very framework Europe built and was mocked for. A genuine vindication of the European approach. But the rules OpenAI favours are capability-based, which critics warn could entrench the dominant labs. Take the vindication, then read the fine print.
Mistral raised €3 billion — the largest equity round in European tech history — to build sovereign AI on European soil. After a month of arguing that Europe regulates but doesn’t build, here’s the counter-evidence. Genuinely good news, and, against OpenAI’s €852 billion, still a minnow. The gap, not the record, is the real story.
Meta launched Muse, an AI agent that doesn’t just answer but acts — sending emails, booking travel, spending money and negotiating on your behalf, on its own. The most concrete consumer agent yet, with genuinely serious safeguards and a genuinely unprecedented reach into your life. The convenience is real, and so is the price. What it means, especially for Europe.
OpenAI says 10,000 AI agents produced a proof for the Navier-Stokes Millennium Problem in 88 hours. The capability is staggering — but it’s a proposed proof, the Clay Institute still lists the problem as unsolved, OpenAI isn’t claiming the prize, and its independence is already disputed. Be amazed, and wait for the mathematicians. And what it means when science becomes a function of compute.
