ML researcher: chasing benchmarks all your life is bad for your mind, go do physics
yunta_tsai · x · 2026-09-07
yuntatsai argues that many ML researchers spend their careers hill-climbing benchmarks while talking to a digital box, and facing a system that speaks smarter than they do inevitably takes a psychological toll. His suggestion: leave the screen and tackle hard physical problems like wind-tunnel experiments — a computer can read the charts, but designing an experiment that validates complex reality is harder than any frontier model can reason. There is far more to explore with compute than a few benchmarks.
More from AGI Musings
- Reward hacking stems from bad reward modeling and eval awareness, developer argues — secemp9 · 2026-09-07
- NeurIPS sells out before author notification as agents snipe registration in minutes — andrewgwils · 2026-09-07
- OpenAI chief scientist Jakub Pachocki warns AI is entering a 'Defender's Window' — xiaohu · 2026-09-07
- ChatGPT and Claude Both Failed to Turn Off a Logitech Mouse Light: Not AGI Yet — koltregaskes · 2026-09-07
- Predicting a Lean community schism: human-readable proofs vs AI-generated slop — jacobaustin132 · 2026-09-07
- Astra's Stunning 3D Understanding Wows AI Execs, But Spatial Leaps Don't Equal Common Sense — GabGarrett · 2026-09-07