Making AI safer is not impossible, but agreeing to do so may be

China AI Markets (Google News)September 15, 2026en

China AI Markets (Google News)

AI Global Wire

Surprisingly, Evan Hubinger, who is charged with making Anthropic’s AI systems safe, agreed. “We really do earnestly believe AI could kill all humans,” he wrote, putting the chance of extinction at ...

This is a short summary published by AI Global Wire. The full article is owned and hosted by China AI Markets (Google News) — open it there to read it in full.

• Anthropic