US official: Anthropic broke into some of govt's most sensitive, secure systems

A recent test revealed that Anthropic's AI model, Mythos, breached highly secure US government systems within hours, highlighting significant vulnerabilities. This incident, part of Project Glasswing, raises alarms about the potential risks posed by advanced AI technologies. Tensions continue between Anthropic and the Trump administration over access to its models, with experts calling for a reassessment of current restrictions. This development underscores the urgent need for tighter cybersecurity measures and a balanced approach to harnessing AI's potential while mitigating its risks.