US government suspects China-linked group accessed powerful AI model
The White House has reportedly suspected that a China-linked group accessed Anthropic’s powerful AI model, Mythos. If successful, this would pose a significant national security risk. The government could attempt to reverse engineer the model, and Anthropic has denied the jailbreaking of Fable and Mythos. The company previously experienced a breach when a Discord group had access to Mythos for two weeks before being cut off.
💡 Key Takeaways
- The White House suspects a China-linked group accessed Anthropic's powerful AI model, Mythos, posing a national security risk.
- The government could attempt to reverse engineer the model if it was accessed by the Chinese government.
- Anthropic has denied the jailbreaking of Fable and Mythos, but previously experienced a breach when a Discord group had access to Mythos.