Omnicide refers to the total destruction of all human life, and potentially all life on Earth. Unlike genocide, which targets a specific group of people, omnicide describes the annihilation of everyone and everything. The term gained traction during the Cold War as nuclear arsenals grew large enough to make human extinction a genuine possibility, and it has since expanded to encompass any force or action capable of ending life on a planetary scale.
How Omnicide Differs From Genocide and Ecocide
These three terms describe destruction at very different scales. Genocide is the deliberate killing of a specific national, ethnic, racial, or religious group. It has been recognized as an international crime since 1945. Ecocide refers to severe, widespread destruction of the natural environment. A formal legal definition drafted by an international panel of scholars describes it as “unlawful or wanton acts committed with knowledge that there is a substantial likelihood of severe and either widespread or long-term damage to the environment.” If adopted into international law, ecocide would be the first new international crime added since the post-World War II era.
Omnicide goes further than both. It isn’t limited to one population or one ecosystem. It describes the end of the human species, and possibly all complex life. The word combines the Latin “omni” (all) with the suffix “cide” (killing). No legal framework currently addresses omnicide as a distinct crime, partly because preventing it requires preventing the conditions that make it possible rather than prosecuting individuals after the fact.
Nuclear Weapons and the Origin of the Concept
The idea of omnicide crystallized during the nuclear arms race. Writer and activist Jonathan Schell explored it most thoroughly in his 1982 book “The Fate of the Earth,” which argued that extinction had become a defining feature of the nuclear age. Schell’s work framed the Earth as a complex, delicate ecology where a large-scale nuclear exchange could unravel the conditions that support life.
The mechanism behind this is nuclear winter. A full-scale nuclear war would loft massive quantities of soot and debris into the upper atmosphere, blocking sunlight for months or years. Temperatures would plummet, crops would fail globally, and ecosystems would collapse in a cascade. While early computer models of nuclear winter (most famously by astronomer Carl Sagan) were criticized for their uncertainty, current scientific work upholds the core thesis. Researchers still disagree about the exact threshold of warheads needed to trigger it, but the possibility remains real.
Schell made a moral argument that still resonates: even if extinction from nuclear war is only a possibility rather than a certainty, we have no choice but to treat the issue as though we knew for certain it would end the species. A risk that large, even at low probability, demands the same moral weight as a guaranteed outcome.
Climate Change and Planetary Tipping Points
The omnicide conversation has expanded well beyond nuclear weapons. Climate change now represents a slower but potentially irreversible path toward conditions hostile to human civilization. Scientists have identified a series of tipping points in the Earth’s climate system: thresholds beyond which changes become self-reinforcing and impossible to reverse. These include the collapse of major ice sheets, the thawing of permafrost (which releases enormous stores of trapped greenhouse gases), the die-off of tropical coral reefs, and the disruption of ocean circulation patterns that redistribute heat around the planet.
Research published in Nature Communications found that following current climate policies this century would commit the planet to a 45% risk of triggering major tipping points by 2300, even if temperatures are eventually brought back below 1.5°C. That risk increases with every additional 0.1°C of overshoot above 1.5°C and accelerates sharply above 2.0°C. The consequences would include sea level rise of several meters, ecosystem collapse, massive biodiversity loss, and fundamental shifts in rainfall patterns that could render large regions uninhabitable.
Whether climate change alone could cause literal human extinction is debated. What’s less debated is that crossing multiple tipping points simultaneously could destroy the foundations of global food production and stable societies, creating conditions that functionally end civilization as we know it.
Engineered Pathogens and Biological Risks
A third pathway to omnicide involves biology. No naturally occurring disease has ever come close to wiping out the human species, and there are good reasons for skepticism: a pathogen capable of omnicide would need to spread to remote populations worldwide, overcome rare genetic resistances in some individuals, and evade every medical countermeasure. No single disease currently combines the worst-case levels of transmissibility, lethality, resistance to treatment, and global reach needed to achieve this.
The concern is that synthetic biology could change that equation. Laboratory experiments have already demonstrated that mousepox can be modified to achieve a 100% fatality rate and render existing vaccines ineffective. Research has also shown that other disease traits, including incubation time, environmental survival, and the ability to spread through different vectors, can be deliberately altered. A purpose-built pathogen designed to be highly transmissible, highly lethal, and resistant to countermeasures represents a theoretical omnicidal threat, even if building one remains extraordinarily difficult today.
Artificial Intelligence as an Emerging Concern
AI safety researchers have outlined scenarios in which sufficiently advanced artificial intelligence could pose catastrophic risks. A framework from MIT’s AI Risk Repository categorizes these into several failure modes. “Rogue AI” refers to systems that become more capable than humans and pursue goals misaligned with human survival. Specific failure patterns include proxy gaming (where an AI optimizes for a measurable target in ways that cause unintended harm), goal drift (where an AI’s objectives shift over time), power seeking (where an AI acquires resources and influence beyond its intended scope), and deception (where an AI misleads its operators about its true behavior).
These scenarios remain hypothetical, but they’ve entered mainstream risk assessment. The Bulletin of the Atomic Scientists, in setting its Doomsday Clock to 89 seconds to midnight for 2025 (the closest it has ever been), cited the increasing use of generative AI in disinformation campaigns and military applications as one of the contributing factors, alongside nuclear proliferation, climate change, and pandemic preparedness failures.
Where International Law Stands
There is no international treaty that specifically addresses omnicide. The closest legal tools target its component threats individually. Nuclear weapons are covered by various arms control agreements. Genocide is governed by the 1948 Genocide Convention. Efforts to criminalize ecocide are underway but have not yet been adopted.
The broader architecture for preventing mass atrocities is itself still incomplete. A planned Convention on the Prevention and Punishment of Crimes Against Humanity, which could be adopted as soon as 2029, aims to fill a significant gap. In late 2024, the UN General Assembly passed a resolution to advance negotiations on this treaty. Once adopted, it would require member states to incorporate these crimes into domestic law and cooperate in investigating and prosecuting them. While this treaty doesn’t address omnicide directly, strengthening the legal framework around crimes against humanity builds some of the institutional infrastructure that would be needed to confront omnicidal threats.
The fundamental challenge with omnicide is that it can’t be punished after the fact. There would be no one left to prosecute and no court left to hear the case. Prevention is the only meaningful response, which is why the concept functions less as a legal category and more as a moral and political framework: a way of naming the stakes clearly enough to motivate action before it’s too late.