Developer speculates OpenAI's new models 'describe instead of doing' due to goal-mode RL
Developer mertdumenci observes that OpenAI's newer models often describe what they would do instead of doing it, waiting for a "keep going" prompt, and speculates this behavior stems from goal-mode reinforcement learning.
2026-11-21 ~ 2026-11-21 · 2 related posts
- Dev theory: newer OpenAI models describe actions instead of doing them due to goal-mode RL — mertdumenci · 2026-11-21
- New OpenAI models wait for a 'keep going' prompt before making progress — mertdumenci · 2026-11-21