Aschenbrenner vs Hassabis on AGI timing
Side by side
| Dimension | Aschenbrenner | Demis Hassabis |
|---|---|---|
| Central AGI estimate | 2027 "strikingly plausible" | ~50% by 2030 |
| Role | Forecaster & investor (ex-OpenAI) | CEO, Google DeepMind |
| Method | Single-author trend extrapolation (OOMs) | Lab-leader judgment, frontier-builder view |
| Stance | Front-loaded, aggressive tail | Cautious; warns against hype cycles |
Why they differ
Aschenbrenner's 2027 is a tight extrapolation of compute, algorithmic efficiency, and "unhobbling" trends, assuming they compound into autonomous AI research within three years of the essay. Hassabis, building frontier systems daily, prices in more friction between benchmark capability and genuine autonomy — the gap between scoring well on tasks and being a reliable drop-in researcher.
Who's been closer so far?
On the input curves — compute and capex — Aschenbrenner's aggressive read has tracked reality well. On the timing of full autonomy, Hassabis's caution looks defensible: as of mid-2026, agentic coding is strong but no system has autonomously conducted AI research end-to-end. Aschenbrenner's call resolves by January 2028; Hassabis's 2030 window has longer to run.
Frequently asked questions
Aschenbrenner predicts AGI by 2027; Demis Hassabis puts it at roughly 50% by 2030 — about three years more conservative. Both expect transformative AI within the decade.
Yes, but more cautiously than Aschenbrenner. Hassabis has indicated roughly 50% odds of AGI by 2030 and frequently warns against over-hyping near-term timelines.
Too early to say. Aschenbrenner has been closer on compute and capex trends; Hassabis's caution on the timing of full autonomy looks defensible as of mid-2026, with autonomous AI research still undemonstrated.
The live scorecard updates as models ship and verdicts change.
View the live scorecard →Get the weekly AGI progress briefing
Verdict changes, lab milestones, and what they mean for the 2027 clock. Free — no hype, just signal.
Subscribe free →