Anthropic suspends new AI tools over US Government security concerns
6 hours ago
- #AI Security
- #Government Regulation
- #Cybersecurity
- Anthropic suspended Claude Fable 5 and Mythos 5 due to US government security concerns.
- The US government cited a method to jailbreak Fable 5, revealing vulnerabilities.
- Anthropic had touted safeguards but allowed pre-release access, acknowledging the tool's potential danger.
- The suspension raises concerns about limiting AI development and international collaboration.
- Anthropic faces a lawsuit over a government designation as a 'supply chain risk', but usage can continue pending litigation.
- The EU emphasizes technological sovereignty following this incident.
- UK tests showed the model could exploit defenses 73% of the time, highlighting a cybersecurity step change.