Anthropic Limits Mythos Release

Anthropic, the AI startup behind the Claude model, has limited the release of its newest model, Mythos, due to concerns over its potential to find security exploits in software. According to TechCrunch, Anthropic's decision to restrict access to Mythos may be a precautionary measure to protect the internet from potential cyber threats.

Mythos has been described as the most psychologically settled model trained by Anthropic to date. However, its capabilities have also raised concerns among experts, who fear that it could be used as a hacker's superweapon.

The limited release of Mythos has already sparked high-level meetings, with Treasury Secretary Scott Bessent and Fed Chair Jerome Powell convening an emergency meeting with Wall Street CEOs to discuss the potential cyber risks associated with the model.