Debating Sentience and Moral Status in AI Training

Scott Alexander and David Manheim debate whether negative reinforcement amounts to punishment and if models can suffer during training, while others question at what point between random initialization and full fine-tuning an LLM deserves moral consideration.

2026-08-31 ~ 2026-09-01 · 2 related posts

• Moral Status of LLM Weights: When Does a Model Deserve Consideration? — davidmanheim · 2026-08-31
• Negative Reinforcement vs. Punishment: Scott Alexander on AI Alignment and Sentience — davidmanheim · 2026-09-01