A hallucination generated by an AI model nearly triggered a US military operation, according to TechCrunch. The incident underscores the risks of deploying large language models in national security settings, where fabricated information could lead to real-world action. A GovAI research scholar warn