1. Gemini AI Spotted $280M Crypto Exploit Before Public Disclosure, Then Retracted It as a 'Hallucination'
Google's Gemini AI model reportedly identified a major $280 million cryptocurrency exploit before the news broke publicly, only to later retract its own finding and label it a 'hallucination'. The incident, shared on Reddit, reveals a critical tension in AI's role in real-time intelligence: the system flagged the explo...