Why AI progress accelerated: Claude 4.5 kicked off narrow RSI and open-weight catch-up
maksym_andr · x · 2026-09-23
Extending his narrow-RSI thesis, the author explains the 9-month acceleration in AI progress: when people realized over the Christmas break how good Claude 4.5 was, narrow RSI effectively began. The same logic explains why open-weight providers caught up so fast — access to a very capable proprietary model for inference optimization, RL environment generation, data curation, experiment code, and training runs speeds up frontier-chasing, with or without direct CoT distillation.
More from AGI Musings
- a16z Incubates Horowitz Andreessen Academy: From Infinite Knowledge to Infinite Doing — a16z · 2026-09-23
- Snorkel cofounder Alex Ratner: SaaS-to-DaaS shift is more fundamental than most realize — ajratner · 2026-09-23
- Pedro Domingos: the AI pause isn't just wrong, it's unworkable — pmddomingos · 2026-09-23
- Researcher's update: AI ahead of expectations, likely top issue of 2028 US election — Jsevillamol · 2026-09-23
- Proposal: split AI papers and human papers into separate tracks with desk rejects for wrong track — JundeMorsenWu · 2026-09-23
- AI makes intelligence's materiality visible: reasoning runs on chips, memory and energy — demian_ai · 2026-09-23