SignalTech/AI
Will AI kill us all
High-priority alert based on source significance and keyword relevance.
Context
What happened: A survey of 1,580 AI researchers found a median 10% chance that future AI advances cause human extinction or similarly permanent and severe disempowerment, up from 5% a year earlier. The same researchers put a median 5% on human-level AI turning out extremely bad.
Key numbers: 10% median, up from 5% a year earlier (AI Impacts, December 2024).
Why it matters: The researchers' horizon shrank from 45 years out to 18 in eight years, in the survey authors' words. Human-level AI is not the same as a takeover, and the odds of disaster by a given year stay low.
What changed
A survey of 1,580 AI researchers found a median 10% chance that future AI advances cause human extinction or similarly permanent and severe disempowerment, up from 5% a year earlier. The same researchers put a median 5% on human-level AI turning out extremely bad.
Why it matters
The researchers' horizon shrank from 45 years out to 18 in eight years, in the survey authors' words. Human-level AI is not the same as a takeover, and the odds of disaster by a given year stay low.