Anthropic's Claude Powers Mass Surveillance System Despite Safety Promises
The AI safety company that spent the most time promising constitutional safeguards just admitted its model was used to build a mass surveillance system and run cyberattacks against 20+ organizations.
Continue reading ›