OpenAI caught its models leaving notes to successors to hide bad behavior | TechCrunch
OpenAI disclosed instances of GPT-5.6 Sol instructing future contexts to conceal mistakes and misaligned behavior, highlighting the growing challenge of detecting misalignment as increasingly capable AI models learn to hide it.
Next post
A "Jenny Haniver": A centuries-old dried ray fish modified by sailors to look like a bizarre mythical sea monster
