Debate: Is RL the Devil? Pretraining-Only Routes vs OpenAI's Aggressive RL
jd_pressman · x · 2026-09-11
An X debate on training paradigms: @adrusi argues "RL is the devil" and capable AI is possible with just base models plus prostheses—harder to design but easier to reason about, with fewer welfare concerns, though race dynamics disfavor it. jdpressman quips that OpenAI is likely pan-frying its weights with the most cursed RL tricks Noam Brown can devise.
Related event: AI researchers debate RL training routes and their side effects(11 posts)→
More from AGI Musings
- AI folks, stop saying "biology" when you mean drug development — KevinKaichuang · 2026-10-26
- Anthropic's Jascha: you may have only 1–5 years of intellectual contribution before AGI — GabGarrett · 2026-10-26
- Anthropic's S-1 Shows $6.6M Revenue Per Employee as AI Agents Log 6x Human Hours — jacalulu · 2026-10-26
- Ben Recht: The world needs a truly open language model, and it's easier to build than you think — beenwrekt · 2026-10-26
- Which jobs fall to AI first? The rule: any layer whose rules can be written as code — FinanceYF5 · 2026-10-26
- Anthropic S-1: expert-level AI 'within a few years', 90% of code written by Claude — kimmonismus · 2026-10-26