Inside Anthropic's Quest to Instill Morality into Its AI Models
bookofjoe · hn · 2026-10-04
A New York Times feature examines how Anthropic works to instill moral judgment into Claude: constitutional-style principles, ethics-focused training, and red-teaming shape the model's values and refusal boundaries. The piece explores the technical approaches and the thorny question of who decides what the model considers right and wrong.
More from Models
- Developers praise Gemini 4 Neon: not the smartest model, but the best to work with — soumitrashukla9 · 2026-11-20
- QuixiInfer: from-scratch Rust engine runs DeepSeek with 256k context on a single RTX 5090 at ~30 tok/s — QuixiAI · 2026-11-20
- Matthew Green questions why a compromised inference provider served an abliterated model — matthew_d_green · 2026-11-20
- GPT 6.2 Sol remains #1 for computer/browser use, says Matthew Berman — MatthewBerman · 2026-11-20
- Microsoft's MAI models have been front-page on Hacker News since Ignite debut — marlene_zw · 2026-11-20
- Gemini 4 Carbon tops graduate-level physics benchmark but loses on agentic tasks — ArtificialAnlys · 2026-11-20