AI news · Research · risk
Lancet panel puts malicious AI among threats to human survival
A 35-member commission ranks AI misuse, including AI-designed bioweapons, and nuclear war as the two risks that could end humanity by 2100.
Malicious use of AI and nuclear war are the two threats that could make humans extinct by 2100, according to the Lancet Commission on 21st Century Global Threats to Health, whose findings the Guardian reported on Sunday 11 October 2026. Of the 17 health threats its 35 experts weighed up to 2100, those two rank first.
The commission rates the likelihood of either as low. Its AI case centres on biological weapons made with AI; its nuclear case on soot from a war blocking sunlight and causing famine. Below them sit 15 “major catastrophic threats”, each of which could cost more than a billion years of healthy life, the Guardian reports, among them climate breakdown, antimicrobial resistance, natural pandemics and cuts to foreign aid. A threat only made the list if policy can reduce it.
The ranking counts disability-adjusted life years from 2025 to 2100, measured as the gap between a plausible best and worst case for each threat, according to the World Health Summit in Berlin, where co-chair Natalia Kanem chairs the launch on Sunday, co-hosted by The Lancet and the Institute for Health Metrics and Evaluation. The commission says it will update the list every year, the Guardian adds.
The report gives health ministries a ranked, measured case for treating AI misuse as a public health matter.
Sources
- The Guardian: Nuclear war and malicious use of AI threaten end of humanity by 2100, says Lancet assessment. Checked
- World Health Summit: Launch of the Findings from the Lancet Commission on 21st Century Global Threats to Health. Checked
Alexandre S.
Alexandre S. is a creative ad strategist and copywriter with ten years' experience in digital marketing across multiple channels.