{"id":"daniel-kokotajlo","name":"Daniel Kokotajlo","kind":"person","aliases":[],"summary":"Daniel Kokotajlo develops AI forecasts and scenarios, including AI 2027; scenario assumptions and revisions remain distinct from observed outcomes.","url":"https://ai-2027.com/","referenceUrls":["https://ai-2027.com/","https://www.dwarkesh.com/p/timelines-june-2025"],"sourceIds":["daniel-kokotajlo-reference"],"topicIds":["intelligence-and-discovery","work-and-agency"],"version":"entity-catalog-v1","path":"/entities/daniel-kokotajlo","json":"/entities/daniel-kokotajlo.json","markdown":"/entities/daniel-kokotajlo.md","relationships":{"sourceIds":["daniel-kokotajlo-reference"],"noteIds":[],"perspectiveIds":[],"parallelIds":[],"topicIds":["intelligence-and-discovery","work-and-agency"]},"publications":"/api/pipeline/publications?entity=daniel-kokotajlo","profile":{"schemaVersion":1,"identityId":"daniel-kokotajlo","tracked":{"id":"daniel-kokotajlo","name":"Daniel Kokotajlo","role":"AI forecaster and AI 2027 coauthor","roles":["researcher","analyst"],"whyTracked":"Produces explicit AI scenarios and engages directly with disputed learning timelines.","scope":"AI progress and research automation; scenario writing is not an observed outcome.","interests":"Coauthors AI 2027; his published forecasts are part of the argument being examined.","eligibilitySources":[{"title":"AI 2027 authors and research","url":"https://ai-2027.com/"},{"title":"His dated reply to Patel","url":"https://www.dwarkesh.com/p/timelines-june-2025"}],"reviewedAt":"2026-10-14"},"contributorContexts":[],"affiliations":[{"personId":"daniel-kokotajlo","personName":"Daniel Kokotajlo","organizationId":"ai-futures-project","organizationName":"AI Futures Project","role":"AI Futures Project coauthor, stating his own median","date":"2026-04-02","dateKind":"source-publication","sources":[{"title":"Q1 2026 Timelines Update","url":"https://blog.aifutures.org/p/q1-2026-timelines-update"}]}],"statements":[{"id":"kokotajlo-automated-coder-2026","questionId":"ai-and-work","personId":"daniel-kokotajlo","kind":"forecast","title":"Automated coding staff by mid-2028","summary":"Kokotajlo's median for the Automated Coder milestone moved from late 2029 to mid-2028. The milestone is the point at which an AGI company would rather lay off all its human software engineers than stop using AI for software engineering.","groupId":"capability-and-diffusion","topicIds":["task-automation","job-displacement-and-creation"],"conceptIds":["task-versus-job"],"roleAtSource":"AI Futures Project coauthor, stating his own median","source":{"title":"Q1 2026 Timelines Update","publisher":"AI Futures Project","url":"https://blog.aifutures.org/p/q1-2026-timelines-update","publishedAt":"2026-04-02","dateBasis":"Byline date reads Apr 02, 2026.","locator":"Opening paragraphs stating Daniel's Automated Coder median and defining the milestone","accessNote":"Post text retrieved. The median sentence is specifically Daniel's. Eli Lifland's median is stated separately in the same post."},"attribution":{"kind":"coauthor","creditedTo":"Daniel Kokotajlo, in a post with Eli Lifland and Brendan Halstead"},"conditions":["The milestone is replacement of human software engineers at an AGI company, not learning across all white-collar work.","The post attributes the shorter median partly to a faster coding-time-horizon trend and to lowering his time-horizon requirement after Opus 4.6."],"limitations":["Late 2029 is the previous median as reported in this post; the earlier source of that figure was not checked in this review.","Mid-2028 is not converted into a month. A median is recorded as 50% because he called it a median."],"support":[],"challenge":[],"evidence":[],"editorialAsOf":"2026-10-14","comparisonNote":"Automated Coder is not the end-2028 humanlike-learning milestone in his June 2025 reply, and not an experienced-editor forecast.","forecast":{"endpointId":"automated-coder-agi-company","endpointLabel":"An AGI company would rather lay off all its human software engineers than stop using AI for software engineering","target":{"startYear":2028,"endYear":2028},"timeWording":"median mid-2028","probability":0.5,"conditions":["Coding labor at an AGI company, under the post's definition."],"targetBasis":"explicit-year","targetDerivation":"The post states the median as mid-2028. The half-year word is kept and not turned into June.","resolutionNote":"His stated median. The post gives no separate scoring protocol beyond the layoff definition."}},{"id":"kokotajlo-learning-2028","questionId":"ai-and-work","personId":"daniel-kokotajlo","kind":"forecast","title":"AI research could shorten the wait","summary":"Kokotajlo gives end-2028 as his median for the learning milestone, anticipating faster progress after automated coding and research.","groupId":"accelerating-learning","topicIds":["continual-learning","job-displacement-and-creation"],"conceptIds":["continual-learning","task-versus-job"],"roleAtSource":"AI forecaster responding to Patel","source":{"title":"Why I don't think AGI is right around the corner","publisher":"Dwarkesh Podcast","url":"https://www.dwarkesh.com/p/timelines-june-2025","publishedAt":"2025-06-03","locator":"Daniel Kokotajlo's dated comment below the essay","dateBasis":"Date of this author's comment, not the parent essay's date.","accessNote":"Original text reviewed; audio/video not independently verified. The link may reflect publisher edits after publication."},"attribution":{"kind":"comment-author","creditedTo":"Daniel Kokotajlo"},"conditions":["Faster AI research is a central assumption."],"limitations":["A forecast in a direct reply, not a demonstrated learning system."],"evidence":[{"title":"The forecast being challenged","url":"https://www.dwarkesh.com/p/timelines-june-2025","publishedAt":"2025-06-02","relationship":"challenge","note":"Patel expects this milestone later."}],"comparisonNote":"His reply explicitly identifies Patel's continual-learning milestone. It is not an independent replication of Patel's observations.","forecast":{"endpointId":"humanlike-learning-white-collar","endpointLabel":"Learning on the job like a human across white-collar work","target":{"startYear":2028,"endYear":2028},"timeWording":"50% by the end of 2028","probability":0.5,"conditions":["Automation accelerates the research needed for flexible learning."],"targetBasis":"explicit-year","targetDerivation":"Year and probability explicitly stated in the reply.","resolutionNote":"No operational scoring protocol agreed; this remains unresolved."},"editorialAsOf":"2026-10-14","support":[],"challenge":[]}],"questions":[{"id":"ai-and-work","title":"When does an AI become a dependable colleague?","shortTitle":"AI & work","summary":"People disagree about which abilities still separate useful assistance from taking responsibility for a whole job. Learning, deployment and the cost of checking all matter.","topicIds":["work-and-business","continual-learning","job-displacement-and-creation"],"conceptIds":["continual-learning","task-versus-job","productivity"],"groups":[{"id":"learning-bottleneck","label":"Learning is the bottleneck","description":"Better performance must carry across real work and feedback."},{"id":"accelerating-learning","label":"Research may accelerate learning","description":"Automated AI research could shorten the route to learning on the job."},{"id":"capability-and-diffusion","label":"Capability can arrive before adoption","description":"General abilities and context may get there; deployment still takes time."},{"id":"labor-transition","label":"Watch the labor market","description":"Interpreting economic signals requires separating correlation from cause."}],"editorialAsOf":"2026-10-14","coverageNote":"Selected historical and recent statements. Workforce share, cashier displacement, learning, editor and Automated Coder forecasts have different endpoints; only the explicitly shared learning endpoint is compared directly. No outcomes have been scored."}],"readings":[],"relations":[],"boundary":"Escape Meridian selects and paraphrases these records. A mention, shared topic or affiliation does not establish authorship, endorsement, agreement or causation. Missing positions remain unknown."},"currentReadings":{"entityId":"daniel-kokotajlo","items":[],"scanned":12,"truncated":false,"scanLimit":500,"checkedAt":"2026-10-18T04:00:18.012Z"},"currentReadingsUrl":"/entities/daniel-kokotajlo.json"}