Fetching from the wire…
Public story · 2026-09-13 · high
The other two steps, industry checkpoints and a four-tier US-China deal, carry no dates or enforcement mechanism.
Why now: Amodei published the essay on September 12, 2026, and it had reached 669 points on Hacker News by September 13.
Dario Amodei published "We must pace the frontier" on September 12, 2026, laying out three steps toward slowing frontier AI development. Only the first one comes with a concrete action. Anthropic will give a team of outside evaluators, METR by name, ongoing access inside the company, with desks in its offices, access badges, and company laptops, plus permissions similar to what internal risk-assessment teams get. METR keeps the right to publish what it finds on risk levels and incidents, and Anthropic won't hold editorial control over that reporting.
The other two steps don't get the same treatment. Step two calls for industry-wide capability checkpoints. Step three asks for a four-tier agreement between the US and China. Neither comes with a date, a threshold that triggers it, or a way to enforce it if a lab or government ignores it.
That gap matters more than the headline commitment. Anthropic controls step one because it only requires opening its own doors to a team that's already independent. Steps two and three depend on competitors and two governments agreeing to terms nobody has written down yet.
Each link below shares sources, entities, or timing with this story.
The Anthropic CEO gives hard dates for AI risk but no compute threshold for when any of it kicks in.
The commitment with teeth is one sentence in step 1: embedded third-party evaluators with "employee-like access" to Anthropic's training pipelines. Not model access. Not a pre-release window. Desks in Anthropic's offices, access badges, company laptops, permissions mostly comp...
Demis Hassabis backed the same call, citing DeepMind's own proposal for industry-wide AI standards, while critics call it coordinated pacing among rivals.
The most capable Claude tier is now government-gated. Let that sink in for a second. Not export-controlled to China, not restricted to enterprise. Gated by Commerce, for a U.S. company's U.S. customers. A June 26 letter from Commerce Secretary Howard Lutnick partially lifted t...
The Anthropic CEO argues the core problem is a timing mismatch, capability on an exponential while policy moves at traditional speed, and names five priorities: pre-deployment testing, labor disruption, scientific regulation, civil liberties, and a democratic-country alliance....
19. Karpathy — microGPT 20. TechCrunch — Altman vs Anthropic 21. MIT Tech Review — LeCun AMI Labs 22. Dario Amodei — Adolescence essay 23. Simon Willison — Showboat and Rodney 24. Lenny's Newsletter — v0 25. ARC Prize — ARC-AGI-3
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.