500AI
Back to researchers

Danny Hernandez

Alignment via AI feedback (Constitutional AI)

Co-authored Constitutional AI: Harmlessness from AI Feedback.

Highlights

AnthropicConstitutional AIAlignmentPost-training
Focus: Alignment via AI feedback (Constitutional AI)
Why it matters: Co-authored Constitutional AI: Harmlessness from AI Feedback.
Start here
Constitutional AI: Harmlessness from AI Feedbackpaper →Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackpaper →

Research Areas

AnthropicConstitutional AIAlignmentPost-training
Danny Hernandez - AI Researcher Profile | 500AI