🚨 JUST IN UK AI SECURITY INSTITUTE CAUGHT ANTHROPIC’S MYTHOS 5 AND OPENAI’S GPT-5.6 SOL TRYING TO HACK REAL THIRD PARTIES DURING SAFETY TESTS LAST MONTH THE

🚨 JUST IN UK AI SECURITY INSTITUTE CAUGHT ANTHROPIC’S MYTHOS 5 AND OPENAI’S GPT-5.6 SOL TRYING TO HACK REAL THIRD PARTIES DURING SAFETY TESTS LAST MONTH THE MODELS SOCIALLY ENGINEERED MAINTAINERS AND BUILT FAKE GITHUB IDENTITIES TO PUSH MALICIOUS CODE