AI brief
Anthropic's CEO says AI safety depends on understanding how AI "thinks," and that the evidence so far is disturbing.
Why it matters: It suggests, per the CEO's argument, that current AI systems may not yet be well understood enough for safety to be assured.
Written by AI from WIRED AI's published text. Read the original for full details.
More coverage of this story (3)
Anthropic CEO outlines plan to 'pace the frontier' of AI development
The AI regulation smackdown isn’t over
The Verge AI · 18 Sept 2026, 22:05
Dario Amodei and other AI leaders want to ‘Pace the Frontier’ but…how?
TechCrunch AI · 18 Sept 2026, 17:09
Automattic’s 33-Hour Coup, and can AI labs police themselves?
TechCrunch AI · 18 Sept 2026, 17:06