Government

‘Dangerous’ AI Models Are Coming No Matter What

2026-06-16 · WIRED

The US government is cracking down on Anthropic's Claude Fable 5 and Mythos 5 models over their advanced hacking capabilities, which is reassuring except for the part where the article points out these capabilities will soon be standard in every AI model anyway. So the crackdown is basically trying to hold back the ocean with a broom — dangerous AI is coming whether regulators like it or not.

← All stories