Person
Ajeya Cotra
Ajeya Cotra is an AI-forecasting researcher best known for the "biological anchors" report on AI timelines written at Open Philanthropy, work that shaped how a generation of researchers argued about when transformative AI might arrive. She co-writes Planned Obsolescence, a blog explaining alignment and AI-risk thinking in plain language.
Appears alongside
Featured in threads
Tracks
- Safety & alignment 1
- Security & misuse 1
Commentary by Ajeya Cotra
From the commentary rail — every link leaves the site for the original piece.
- 25 September 2026 · Planned ObsolescenceEvidence about risk should be transparentWe can’t develop safety standards if we have to rely on opaque judgment
- 28 August 2026 · Planned ObsolescenceThe Hugging Face attack surprised meIt’s a major warning shot, and might be the last one we get
- 14 August 2026 · Planned ObsolescenceHurtling through 2026Scoring ten qualitative forecasts five months early
- 10 July 2026 · Planned ObsolescenceTotal research transparency would be niceWe could learn what's scary and stop doing that
- 5 June 2026 · Planned ObsolescenceCould a company overpower nations?The leading AI company could gain unprecedented power
- 1 May 2026 · Planned ObsolescenceScience and speculationWill we agree about AI risks in time?
- 3 April 2026 · Planned ObsolescenceSix milestones for AI automationWhat can AI do on its own, and how well?
- 5 March 2026 · Planned ObsolescenceI underestimated AI capabilities (again)Revisiting a prediction ten months early
- 14 January 2026 · Planned ObsolescenceAI predictions for 2026But first, scoring my predictions for 2025
- 6 January 2026 · Planned ObsolescenceSelf-sufficient AINo, we don't "have AGI already." But in any case, we should articulate clearer milestones.
- 10 October 2023 · Planned ObsolescenceScale, schlep, and systemsThis startlingly fast progress in LLMs was driven both by scaling up LLMs and doing schlep to make usable systems out of them. We think scale and schlep will both improve rapidly.
- 29 August 2023 · Planned ObsolescenceLanguage models surprised usMost experts were surprised by progress in language models in 2022 and 2023. There may be more surprises ahead, so experts should register their forecasts now about 2024 and 2025.
- 4 April 2023 · Planned ObsolescenceAIs accelerating AI researchResearchers could potentially design the next generation of ML models more quickly by delegating some work to existing models, creating a feedback loop of ever-accelerating progress.
- 26 March 2023 · Planned Obsolescence"Aligned" shouldn't be a synonym for "good"Perfect alignment just means that AI systems won’t want to deliberately disregard their designers' intent; it's not enough to ensure AI is good for the world.
- 26 March 2023 · Planned ObsolescenceAlignment researchers disagree a lotMany fellow alignment researchers may be operating under radically different assumptions from you.
- 26 March 2023 · Planned ObsolescenceTraining AIs to help us align AIsIf we can accurately recognize good performance on alignment, we could elicit lots of useful alignment work from our models, even if they're playing the training game.