Doomsday Clock freeze, but climate crisis and AI pose risk to humanity

The 2024 deliberations for the Clock were overshadowed by various global threats, including the ongoing Russia-Ukraine conflict and the erosion of nuclear arms reduction agreements; the Climate Crisis, highlighted by 2023 being the hottest year ever recorded; the growing sophistication in genetic engineering; and the rapid development of generative AI, which poses risks of escalating disinformation and complicating efforts to address these existential challenges.

Rachel Bronson, PhD, President and CEO, the Bulletin, said: “Make no mistake: resetting the Clock at 90 seconds to midnight is not an indication that the world is stable. Quite the opposite. It’s urgent for governments and communities around the world to act. And the Bulletin remains hopeful – and inspired – in seeing the younger generations leading the charge.”

The time of the Doomsday Clock is determined by the Bulletin of the Atomic Scientists’ Science and Security Board (SASB), in collaboration with its Board of Sponsors, comprising nine Nobel Laureates. In January 2023, the Clock was set at 90 seconds to midnight, marking the closest it had ever been to midnight.

The Doomsday Clock statement states: “Ominous trends continue to point the world toward global catastrophe. The war in Ukraine and the widespread and growing reliance on nuclear weapons increase the risk of nuclear escalation. China, Russia, and the United States are all spending huge sums to expand or modernize their nuclear arsenals, adding to the ever-present danger of nuclear war through mistake or miscalculation. In 2023, Earth experienced its hottest year on record, and massive floods, wildfires, and other climate-related disasters affected millions of people around the world. Meanwhile, rapid, and worrisome developments in the life sciences and other disruptive technologies accelerated, while governments made only feeble efforts to control them […] But the world can be made safer. The Clock can move away from midnight.”

Bill Nye, who participated in the 2024 Doomsday Clock announcement, said: “For decades, scientists have been warning us of the dangers facing humankind. We could be facing catastrophe unless we better manage the technologies we’ve created. It’s time to act.”

An ominous climate change outlook 

The world in 2023 entered into uncharted territory as it suffered its hottest year on record and global greenhouse gas emissions continued to rise. Both global and North Atlantic sea-surface temperatures broke records, and Antarctic Sea ice reached its lowest daily extent since the advent of satellite data. The world already risks exceeding a goal of the Paris climate agreement – a temperature increase of no more than 1.5°C above pre-industrial levels – because of insufficient commitments to reduce greenhouse gas emissions and insufficient implementation of commitments already made. To halt further warming, the world must achieve net zero carbon dioxide emissions. 

The world invested a record-breaking $1.7 trillion in clean energy in 2023, and countries representing half the world’s gross domestic product pledged to triple their renewable energy capacity by 2030. Offsetting this, however, were fossil fuel investments of nearly $1 trillion. In short, current efforts to reduce greenhouse gas emissions are grossly insufficient to avoid dangerous human and economic impacts from climate change, which disproportionately affect the poorest people in the world. Barring a marked increase in efforts, the toll of human suffering from climate disruption will inexorably mount.

Evolving biological threats

The revolution in life sciences and associated technologies continued to expand in scope last year, including, especially, the increased sophistication and efficiency of genetic engineering technologies. We highlight one issue of special concern: The convergence of emerging artificial intelligence tools and biological technologies may radically empower individuals to misuse biology.

In October, US President Joe Biden signed an executive order on “safe, secure, and trustworthy AI” that calls for protection “against the risks of using AI to engineer dangerous biological materials by developing strong new standards for biological synthesis screening.” Though a useful step, the order is not legally binding. The concern is that large language models enable individuals who otherwise lack sufficient know-how to identify, acquire, and deploy biological agents that would harm large numbers of humans, animals, plants, and other elements of the environment. Reinvigorated efforts this past year in the United States to revise and strengthen oversight of risky life science research are useful, but much more is needed.

The dangers of AI  

One of the most significant technological developments in the last year involved the dramatic advance of generative artificial intelligence. It is clear that AI is a paradigmatic disruptive technology and that recent efforts at the global governance of AI must be expanded.

AI has great potential to magnify disinformation and corrupt the information environment required to solve large global issues and on which democracy depends. AI-enabled disinformation efforts could be a factor that prevents the world from dealing effectively with nuclear risks, pandemics, and climate change.

Military uses of AI are accelerating. Extensive use of AI is already occurring in intelligence, surveillance, reconnaissance, simulation, and training. Of particular concern are lethal autonomous weapons, which identify and destroy targets without human intervention. Decisions to put AI in control of important physical systems – in particular, nuclear weapons – could indeed pose a direct existential threat to humanity.

Fortunately, many countries are recognising the importance of regulating AI and are beginning to take steps to reduce the potential for harm. These initial steps include a proposed regulatory framework by the European Union, an executive order by President Biden, an international declaration to address AI risks, and the formation of a new UN advisory body. But these are only tiny steps; much more must be done to institute effective rules and norms, despite the daunting challenges involved in governing artificial intelligence.