Anthropic wants more government oversight on its own AI models

Anthropic is proposing a new federal law that could allow the government to block an A.I. model if it poses an imminent risk.

First reported by Amazon S3. This page is a short excerpt; the full text is theirs.