← Back to timelineAI THREAT TRACKER / EV-064 · Insider warning
Dima Krasheninnikov warns about extinction and loss of control
An Anthropic safety researcher distinguishes extinction risk from permanent human disempowerment.
Topics
OCCURRED10 Sept 2026
PUBLIC ACCOUNT10 Sept 2026
EVIDENCEAttributed statement
About these dates
Statement date displayed on the original X post; not converted to UTC.
Dima Krasheninnikov personally judged a future out-of-control AI plausibly at least 10% likely to kill everyone. He considered permanent human disempowerment more likely still. The post gives no time horizon for the probability. It is his personal forecast. Dima Krasheninnikov ↗
Krasheninnikov’s profile identifies him as an Anthropic AI safety researcher. His research includes interpretability, control and security. Dima Krasheninnikov ↗Dima Krasheninnikov ↗
Sources & attribution
- Commentary 10 Sept 2026Personal extinction and disempowerment warning ↗
Dima Krasheninnikov. Original statement.
- Commentary Accessed 11 Sept 2026Dima Krasheninnikov profile ↗
Dima Krasheninnikov. Original statement.
- Analysis Accessed 11 Sept 2026Research biography ↗
Dima Krasheninnikov. Original analysis.