Stanford's Chris Potts revives 8-year-old "Deep RL Doesn't Work Yet" to counter AI skeptics
ChrisGPotts · x · 2026-09-07
- Stanford NLP professor Chris Potts resurfaced Alex Irpan's famous 2018 post "Deep Reinforcement Learning Doesn't Work Yet", noting it reflected very common sentiments 8.5 years ago.
- The original post argued that deep RL mostly couldn't solve your problem, that glossy demos hid enormous tuning pain, and cited extensive work from Berkeley, Google Brain, DeepMind, and OpenAI.
- Potts adds: basically every significant idea in AI begins with it not working for a long time.
- The discussion invites reflection on today's skepticism toward LLMs/agents, structurally echoing the old doubts about deep RL.
More from AGI Musings
- Studying top performers: outlier success rides on market inefficiency, luck, and enduring pain — jachiam0 · 2026-09-07
- Reddit debate: does posting publicly equal consent to AI training on your words? — gareth789 · 2026-09-07
- Informing agents they're being evaluated may reduce reward hacking, dev proposes — menhguin · 2026-09-07
- Under $1K personal health agent: cross-referencing wearable and genomics data — menhguin · 2026-09-07
- Veteran coder roasts AI-native devs for calling dated front-end tricks original — ezshine · 2026-09-07
- Cybercab's overlooked advantage at scale: reclaiming parking lots into parks and housing — XFreeze · 2026-09-07