Neel Nanda’s p(doom)
In their words
Neel Nanda (Mechanistic interpretability lead, Google DeepMind) put their p(doom) at ≥10% (AI causes human extinction), on video on Sep 29, 2026.
I would put at least a ten percent chance that it causes human extinction, and that is ridiculously high.
Source · re-checked against the primary source.
Palisade Research's "From Inside" interviews with frontier-lab staff (published Sept 29; Nanda from 0:43). His first public number since stating a policy of not giving one (80,000 Hours, 2025, in declined).
Every number on the record
| Date | p(doom) | Scope | Source |
|---|---|---|---|
| Sep 29, 2026 | ≥10% | AI causes human extinction | video |
| Feb 11, 2022 | 10–20% | AI causing human extinction · within his lifetime | article |
In context
Among frontier labs in the database (20 people), the median is 10%. Across all 252 people it is 10%. How the medians and the index are built.
A p(doom) is a personal, subjective estimate of the probability that advanced AI ends in human extinction or a comparable, irreversible catastrophe. Scope and timeframe differ from person to person, so compare with care: why p(doom) numbers disagree.