Person
Anders Cairns Woodruff
Commentary by Anders Cairns Woodruff
From the commentary rail — every link leaves the site for the original piece.
- 10 June 2026 · Redwood ResearchEstimating No-CoT Task-Completion Time Horizons of Frontier AI ModelsModels' no-CoT time horizon has doubled roughly every year.
- 11 May 2026 · Redwood ResearchHow useful is the information you get from working inside an AI company?My median guess: it's as good as a crystal ball that sees 2.5 months into the future.
- 27 April 2026 · Redwood ResearchFail safe(r) at alignment by channeling reward-hacking into a "spillway" motivationA controlled reward-seeking motivation could make AI safer and more useful
- 27 March 2026 · Redwood ResearchAI's capability improvements haven't come from it getting less affordableAI inference is still cheap relative to human labor
- 12 March 2026 · Redwood ResearchAre AIs more likely to pursue on-episode or beyond-episode reward?RL would encourage on-episode reward seeking, but beyond-episode reward seekers may learn to goal-guard.
- 26 February 2026 · Redwood ResearchFrontier AI companies probably can't leave the USThe executive branch can, and probably would, block a frontier AI company's departure.