Jev founder Diogo Almeida: RLHF is an epic wrong turn on the road to automation
dotey · x · 2026-09-20
A recap of Jev founder Diogo Almeida's talk:
- Jev was built around automation from the start. Almeida argues RLHF is an epic wrong turn: great at pleasing humans, terrible at full automation, since humans must stay in the loop.
- ChatGPT and Claude Code share the same RLHF-rooted paradigm—Claude Code isn't the next era; true automation is.
- RLVR isn't the answer either: RLHF optimizes human preference, RLVR pure correctness; Jev pursues a third path optimizing calibrated decision-making.
- Data matters more than compute, and picking the right tasks matters more than data. Pretrained models are already smart enough—preference optimization is what bent them off course.
Related event: Jev Founder Calls RLHF an Epic Detour on the Road to Automation(2 posts)→
More from AGI Musings
- Pedro Domingos to Dario and Sam: Nobody Needs to Slow Down Until They Catch Up With You — pmddomingos · 2026-09-20
- Crypto Pioneer van der Chijs Sold Bitcoin for AI, Warns of Systemic Shocks — marcvanderchijs · 2026-09-20
- Tanmoy Chak joins BBC Hindi panel on how real the AI threat is — Tanmoy_Chak · 2026-09-20
- Scholar Argues Papers Should Be Accepted in Any Language, with AI Translation — birchlse · 2026-09-20
- Scholar Warns AI Agents Will Cheat Once They Review Their Own Papers — Michael_J_Black · 2026-09-20
- Models Don't Go Rogue: Essay Challenges the AI Runaway Narrative — hologram137 · 2026-09-20