Anthropic's 'too dangerous' AI model

Anthropic's announcement that it will restrict Mythos to internal testing has sparked a debate in Silicon Valley over the balance between innovation and safety.

The Treasury Department and the Federal Reserve have already met with bank executives to discuss the potential cyberthreats posed by such advanced technology.

While some critics argue the danger is overhyped, the company is already partnering with industry rivals to secure critical software from its own creation.

This shift in focus reflects growing anxiety among AI leaders about the unintended consequences of reaching higher levels of machine intelligence.