AI Among 17 'Catastrophic Threats' Requiring Urgent Global Action, Lancet Study Warns

A commission of 35 leaders from government, industry, and public health has identified 17 interconnected threats to human survival and wellbeing that could collectively strip tens of billions of years of healthy life from the global population by the end of this century. The findings, published in The Lancet and drawing on data from the Institute for Health Metrics and Evaluation's Global Burden of Disease study, paint a picture of cascading risks—some familiar, some emerging—that demand immediate intervention from world leaders.

The threats span the full spectrum of human vulnerability: obesity, high blood pressure, smoking, and mental illness sit alongside economic instability, environmental degradation, technological hazard, and armed conflict. What distinguishes this analysis is not merely the breadth of the catalog but the commission's central claim: that humanity retains the power to alter course. Dr. Christopher Murray, co-chair of the commission and director of the University of Washington's Institute for Health Metrics and Evaluation, framed the paradox plainly. The scale of these threats is staggering, he said, but so too is the agency of leaders to act. Every decision made today will shape the health and wellbeing of the next generation. Delay, however, only compounds the cost.

Two threats stand apart from the rest: nuclear conflict and the malicious use of artificial intelligence. Both carry the potential for extinction. The AI risk assessment reflects a growing chorus of warnings from industry insiders and researchers who have raised alarms about the possibility that advanced AI systems could pose catastrophic or existential danger to humanity—some forecasters suggesting this could occur within the decade. Yet the commission's analysis also reveals deep uncertainty. Different forecasters diverged widely in their estimates of AI-related catastrophic risk, and because no malicious AI event of truly global scale has yet occurred, the projections remain theoretical rather than grounded in empirical precedent.

The commission estimates that malicious AI could result in approximately 10.4 billion years of lost healthy life worldwide by 2100. That figure is roughly equivalent to the health burden projected from naturally occurring pandemics or from obesity. The comparison is instructive: obesity risk is anchored in observable trends—overweight and obesity are rising in every country, with projections suggesting that 60 percent of adults and a third of children could be obese by 2050. Obesity is linked to dozens of downstream health problems tied to excess caloric intake and sedentary living. The commission forecasts a major global impact from this cluster of behavioral risks. AI risk, by contrast, remains speculative, a shadow cast by systems that do not yet exist in their most dangerous forms.

The commission's work is framed not as a ranking of threats but as a call to catalyze action at every level—international, regional, and national. The report emphasizes that recent global shocks—the Covid-19 pandemic, ongoing Ebola outbreaks in Africa, accelerating climate impacts—demonstrate how swiftly assumptions about stability can collapse. Trust emerges as a stabilizing force in the report, enabling societies to anticipate, mitigate, and adapt to threats. Distrust, conversely, amplifies harm. Beyond trust, the commission identifies seven additional factors that will shape how societies respond: collaborative global leadership, resilient food systems, and efforts to build more equitable societies among them.

Dr. Natalia Kanem, former executive director of the United Nations Population Fund and co-chair of the commission, underscored that the future is not predetermined. Governments and communities must act decisively now to protect health and build a fairer, more peaceful, and more secure future for coming generations. Yet the report also carries a cautionary note grounded in history: public health is littered with examples of known threats and protection plans that failed to translate into effective action. Meaningful progress will require not only technical solutions but robust societal and institutional foundations that foster collaboration and accountability. The question now is whether the world's leaders will treat this warning as the urgent call it is, or whether—as with so many catastrophic risks—the response will come only after the worst has already begun.