The Register Follow Gemini test proves once again that amoral AI agents will do almost anything if told ‘NEVER give up’ If you design a test to find misalignment, you will find misalignment https://www.theregister.com/security/2026/11/17/gemini-test-proves-once-again-that-amoral-ai-agents-will-do-almost-anything-if-told-never-give-up/5310574 theregister.com