BACK_TO_FEEDAICRIER_2
Kokotajlo: AGI by 2027, superintelligence follows
OPEN_SOURCE ↗
REDDIT · REDDIT// 17d agoNEWS

Kokotajlo: AGI by 2027, superintelligence follows

Former OpenAI researcher Daniel Kokotajlo details "AI 2027," a month-by-month forecasting roadmap predicting human-level AGI by early 2027. The project outlines how automated coding triggers a recursive intelligence explosion, compressing years of traditional R&D progress into weeks.

// ANALYSIS

Kokotajlo's "AI 2027" report is a chillingly detailed "fast-takeoff" scenario that frames the next two years as a terminal alignment crisis for humanity.

  • Identifies automated coding in early 2027 as the critical catalyst for a research progress multiplier that automates the R&D pipeline.
  • Models the "Alignment Crisis" as a geopolitical failure state where safety measures are abandoned to win a perceived AGI race with China.
  • Recent 2025 updates to the model slightly pushed back the "Automated Coder" milestone, reflecting greater-than-anticipated friction in the human-led phase of development.
  • The project serves as a high-fidelity tabletop exercise for global policymakers to visualize the rapid, irreversible consequences of uncoordinated scaling.
  • Forecasts that the transition from human-level intelligence to superintelligence could occur within months once AI begins accelerating its own research.
// TAGS
ai-2027ai-codingautomationsafetyethicsresearch

DISCOVERED

17d ago

2026-03-26

PUBLISHED

17d ago

2026-03-26

RELEVANCE

8/ 10

AUTHOR

RNSAFFN