Anthropic investigates unauthorized Mythos access by Discord group
Anthropic is investigating unauthorized access to its powerful Mythos AI model by a small group of Discord users. The breach raises concerns about the potential for cyberattacks using the advanced AI model.
Anthropic has launched an investigation into unauthorized access to its Mythos AI model. A small group of Discord users gained access to the model in late February, just as Anthropic began rolling out a preview to select companies. The Mythos model is designed to identify software vulnerabilities, raising fears it could be used to accelerate cyberattacks. Anthropic stated that the breach occurred through a third-party vendor environment and did not impact its systems. Initial claims linking the breach to the ShinyHunters group were debunked as AI-generated fabrications. The incident highlights concerns about controlling access to powerful AI models.
This content was automatically generated and/or translated by AI. It may contain inaccuracies. Please refer to the original sources for verification.