anthropic

Evan Hubinger, a leading AI safety researcher at Anthropic, has expressed concern that advanced AI systems might pose a risk to human survival within the next ten years. This follows reports that Anthropic withheld its latest AI model from a UK safety institute, highlighting ongoing challenges in AI alignment and governance.

Several Claude AI subscribers have reported unauthorized token usage linked to stolen session credentials. Anthropic, the company behind Claude, confirmed that malware-driven theft of login sessions is enabling attackers to consume users' tokens without their knowledge.

Several Claude AI subscribers have reported unauthorized token consumption linked to stolen session credentials. Anthropic confirmed that malware-driven theft of login sessions is enabling attackers to drain user tokens, highlighting gaps in usage transparency and account security.