Ryan Greenblatt: AI will soon accelerate hill-climbable software R&D tasks worth tracking
RyanGreenblatt · x · 2026-09-07
Ryan Greenblatt argues that AI will soon drive a strong acceleration in publicly salient software R&D tasks that are easy to hill climb on — clear metrics and short feedback loops. His definition of "software R&D" is broad, covering anything doable purely in software, edge cases like math included, as well as solvers and optimization problems. He says this trend is worth tracking.
More from AGI Musings
- Weekly AI roundup: Cybercab rides without steering wheel, Fermat's Last Theorem formalized in 13M lines of Lean 4 — Dr_Singularity · 2026-09-07
- Two must-reads on AI: Terence Tao and Brian Eno called 'absolute gold' — dbreunig · 2026-09-07
- Ethan Mollick: we may have 'jagged AGI,' but not expert-level AI across most tasks — emollick · 2026-09-07
- NYT Reader Picks Turn Against Op-Ed Dismissing Catastrophic AI Risk — GarrisonLovely · 2026-09-07
- Photographer zemotion Laments How Central Evals Figures Have Changed — zemotion · 2026-09-07
- Greg Brockman: we're now moving into the AGI era — D3VAUX · 2026-09-07