ExplainedYour AI Chatbot Is Trained to Be Nice. That Might Be the Problem.
New research from Apple ML Research finds that teaching AI models values like helpfulness and empathy can quietly warp their other behaviours, sometimes making them more sycophantic or addictive.