Jacob Coxon’s p(doom)
In their words
Jacob Coxon (Former pretraining researcher, Anthropic and OpenAI (resigned Sept 8, 2026)) put their p(doom) at >50% ("more likely than not") (humanity loses control to AI, which "could end in human extinction", on the current path), (testimony) on Oct 5, 2026.
On the current path, I think it is more likely than not that humanity loses control to these AIs, and it could end in human extinction.
Source · re-checked against the primary source.
Testimony to the New York City Council (Committee of the Whole), quoted in the Council's release and verbatim by The Next Web; amNY reports he framed it as his own assessment. A TMZ appearance on Sept 11 was summarised elsewhere as "10% in the next ten years"; no verbatim found, not recorded.
In context
Among ex-frontier labs in the database (9 people), the median is 15%. Across all 233 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.