Cosmic Extinction: Understanding Humanity’s Potential Self-Destruction

We tend to think of extinction as something that happened to the dinosaurs – a distant event, sealed off from our present lives by 66 million years. But a growing body of philosophical and scientific thought argues otherwise. From the misuse of our own technologies to the indifferent violence of the cosmos, the threats capable of ending the human story are neither as remote nor as abstract as we might like to believe. Canadian philosopher John Leslie, one of the most serious academic voices on this subject, has spent decades arguing that we are almost certainly in danger of extinction – and that we have persistently underestimated the odds. This post explores the dual architecture of that danger: the catastrophes we are building ourselves, and those the universe may deliver uninvited.

John Leslie and the doomsday argument 🔗

John Leslie is University Professor Emeritus at the University of Guelph and a Fellow of the Royal Society of Canada. His 1996 book The End of the World: The Science and Ethics of Human Extinction remains one of the most comprehensive philosophical treatments of the subject. Leslie does not traffic in science fiction. His approach is probabilistic, ethical, and grounded in actual science.

Central to his work is the Doomsday Argument, originally formulated by mathematician Brandon Carter. The argument applies Bayesian reasoning to a simple observation: if the total number of humans who will ever live is relatively small, then we – alive today – should not be surprised to find ourselves near the end of that population. The very fact that we are among the billions already born, rather than the trillions who might yet be born in a long human future, is statistical evidence that the human future may be short.

Leslie stresses that the Doomsday Argument does not generate risk estimates on its own. Rather, as he explains, it acts like a magnifying glass – it scales up whatever probability you already assign to extinction based on the actual dangers in the world. And those dangers, he argues, are serious enough on their own terms. Leslie estimates a 30% chance of human annihilation by nuclear war, disease, or other means within the next 500 years – a figure he considers conservative once the Doomsday Argument’s amplifying effect is applied.

Human-made threats: the dangers we manufacture 🔗

The most philosophically troubling aspect of extinction risk is that much of it is self-inflicted. The same capacities that define human civilization – the ability to engineer, synthesize, and compute at scale – are the very capacities that could undo us. Leslie catalogues a range of such threats, from the familiar to the underappreciated.

Nuclear war and weapons of mass destruction 🔗

Nuclear weapons remain the most visible existential risk humans have created. The discovery of “nuclear winter” in the 1980s demonstrated a specific mechanism by which nuclear war could lead to human extinction, with Carl Sagan arguing that such a conflict would imperil all descendants for as long as humanity endured. Today, the concern is not merely that nuclear weapons exist, but that they remain embedded in geopolitical tensions with no reliable path to elimination.

Biological and chemical threats 🔗

Engineered pathogens occupy a particularly alarming position on the risk spectrum. A RAND analysis found that a combination of designer pathogens could potentially achieve near-total lethality, especially if deployed in a way that prevented communities from isolating themselves. Unlike nuclear weapons, the barriers to creating biological agents have historically been knowledge and expertise rather than expensive equipment – and those barriers are eroding. As researchers at the National Institutes of Health have documented, the rapid evolution of genetic engineering tools makes the deliberate creation of extinction-level pathogens a threat that is no longer purely hypothetical.

Artificial intelligence 🔗

Artificial intelligence represents a newer but increasingly debated vector of existential risk. In 2023, hundreds of AI experts signed a statement declaring that mitigating the risk of extinction from AI should be a global priority alongside pandemics and nuclear war. The concern is not simply that AI systems will “go rogue” in a cinematic sense. It is that sufficiently powerful systems, if misaligned with human values, could pursue goals in ways that are catastrophic and irreversible. As the Brookings Institution notes, the deeper worry is whether, once AI systems improve themselves recursively, they could surpass human oversight capacity entirely.

Even setting aside superintelligence scenarios, experts at the Bulletin of the Atomic Scientists have pointed out that AI could exacerbate existing extinction risks by enabling bad actors to design bioweapons or coordinate attacks on critical infrastructure. The threats are intersectional: AI is not a standalone danger but an accelerant for others.

Environmental collapse 🔗

Climate change and ecological degradation occupy a different temporal register – slower-moving but no less serious. Climate change has initiated biospheric feedback loops capable of destabilizing the planet’s life-support systems, including the biodiversity and ecosystem services on which civilization depends. Leslie includes environmental catastrophe in his survey of extinction scenarios, and subsequent philosophers like Nick Bostrom and Toby Ord have formalized this, defining existential risks as events that would either annihilate intelligent life or permanently foreclose its potential – a category that includes runaway climate change alongside nuclear war and asteroid impact.

Cosmic threats: what the universe may bring 🔗

Beyond human agency lies a different order of danger – threats that originate in the physics of the cosmos itself. Leslie, true to his background in philosophical cosmology, is careful not to limit his account of extinction to things we have made. The universe, he observes, is not designed for our survival.

Supernovae and cosmic radiation 🔗

When a massive star collapses, it releases an amount of energy that can briefly rival the output of billions of stars. The dying star emits high-energy gamma rays and streams of cosmic rays – subatomic particles traveling near the speed of light. A supernova within roughly 30 light-years of Earth would severely deplete the ozone layer, disrupt the marine food chain, and likely cause mass extinction. Fortunately, such proximate events are rare – occurring within that distance only every few hundred million years. But on geological timescales, they are not negligible. Astronomers have found radioactive evidence of a supernova approximately 300 light-years away that exploded 2.5 million years ago, contributing to climate cooling and the extinction of some ancient species.

Gamma-ray bursts and neutron star mergers 🔗

More dangerous still are gamma-ray bursts (GRBs) – the most energetic explosions in the known universe. These events occur when massive stars collapse into black holes, or when two neutron stars merge. In less than a second, a gamma-ray burst can produce more energy than the Sun will emit across its entire 10-billion-year lifetime. The radiation is strongly beamed, meaning that a direct hit on Earth from within our galaxy would sterilize at least one hemisphere of the planet almost instantaneously.

The statistical odds of a gamma-ray burst triggering a mass extinction are estimated at 50% over the past 500 million years and around 90% over the 4 billion years that life has existed on Earth. Some astronomers have argued that the first of the five major mass extinctions – the Late Ordovician event around 440 million years ago, which eliminated two-thirds of all marine species – may have been caused by exactly such a burst. A gamma-ray burst in the Milky Way pointed at Earth remains, according to NASA astrobiologists, one of the most severe natural extinction mechanisms the planet could face.

High-energy physics experiments 🔗

Leslie raises a more unusual cosmic-scale risk: the possibility that experiments in high-energy physics could produce catastrophic and irreversible effects. Particle accelerators recreate conditions close to those of the early universe. In principle – though the probability is considered vanishingly small – a sufficiently energetic collision could trigger a phase transition in the quantum vacuum, releasing energy that would propagate outward at the speed of light, transforming the fundamental laws of physics in its wake. As commentators on Leslie’s work have noted, such an event might already have occurred somewhere in the universe, with consequences racing toward us undetected. Leslie includes this not to induce panic but to illustrate a core philosophical point: some risks are so asymmetric in their consequences that even tiny probabilities demand serious ethical attention.

The ethics of extinction risk 🔗

What makes Leslie’s work philosophically distinctive is its insistence on the ethical weight of extinction. The death of every future person who would ever have been born is not simply a large number of deaths – it is the permanent foreclosure of all human possibility. Philosopher Derek Parfit argued that if Earth remains habitable for a billion more years and can sustain over a billion humans, there is a potential for 10¹⁶ human lives of normal duration – a figure that makes even a small increase in extinction risk morally catastrophic.

Leslie is also alert to what he calls observational selection bias: the fact that we observe an undisastrous past does not mean we are safe. Any observer, by definition, finds themselves in a period before extinction has occurred. This makes historical survival a poor guide to future safety. The question is not whether we have survived so far – we have – but whether the conditions that allowed survival remain in place, and whether we are adding new risks faster than we are mitigating old ones.

As philosopher Nick Bostrom has argued, preventing existential risk is a high moral priority under virtually every coherent ethical framework – consequentialist, deontological, and intuitionist alike. The precise degree of badness may vary between theories, but the conclusion converges: extinction is not just a very large tragedy, it is a categorically different kind of event.

Navigating the precarious balance 🔗

Leslie’s purpose in cataloguing these dangers is not fatalism. His work serves as a cautionary assessment, urging philosophers and policymakers alike to take seriously the risks that future generations will face. The practical implications are clear: global cooperation on arms reduction and pandemic preparedness, investment in AI alignment research, and the development of early warning systems for cosmic threats like near-Earth asteroids and nearby stellar evolution.

What Leslie demands, above all, is a shift in temporal perspective. Our political institutions are organized around election cycles. Our economic systems discount the future. But existential risks operate on timescales of centuries and millennia. RAND researchers have concluded that reducing extinction risk – whether from AI, nuclear weapons, or climate change – requires sustained, coordinated effort that does not treat these threats as competing priorities but as mutually reinforcing challenges. On this point, Leslie would agree entirely.

The philosophy of cosmic extinction is ultimately not about despair. It is about moral seriousness – about recognizing that what is at stake in the decisions made today is not just the welfare of people alive now, but the existence of everyone who might ever come after.

What do you think? If the probability of human extinction within the next 500 years is genuinely significant – as Leslie argues – does that change how we should weigh the interests of future generations against those of people alive today? And given that cosmic risks like gamma-ray bursts are entirely beyond human control, should philosophical and policy discussions of extinction focus primarily on the threats we have created ourselves?

References

1. https://closertotruth.com/contributor/john-leslie/
2. https://www.routledge.com/The-End-of-the-World-The-Science-and-Ethics-of-Human-Extinction/Leslie/p/book/9780415184472
3. https://www.barnesandnoble.com/w/the-end-of-the-world-john-leslie/1128435114
4. https://en.wikipedia.org/wiki/Cosmic_endowment
5. https://www.rand.org/content/dam/rand/pubs/research_reports/RRA3000/RRA3034-1/RAND_RRA3034-1.pdf
6. https://pmc.ncbi.nlm.nih.gov/articles/PMC11116769/
7. https://en.wikipedia.org/wiki/Existential_risk_from_artificial_intelligence
8. https://www.brookings.edu/articles/are-ai-existential-risks-real-and-what-should-we-do-about-them/
9. https://thebulletin.org/2024/07/three-key-misconceptions-in-the-debate-about-ai-and-existential-risk/
10. https://countercurrents.org/2025/12/examining-the-three-existential-threats-of-the-21st-century-artificial-intelligence-climate-change-and-nuclear-weapons/
11. https://link.springer.com/article/10.1007/s13347-024-00706-2
12. https://earthsky.org/space/threat-of-supernovas-to-life-on-earth/
13. https://supernova.eso.org/exhibition/1217/
14. https://astrobiology.nasa.gov/news/how-deadly-would-a-nearby-gamma-ray-burst-be/
15. https://www.goodreads.com/book/show/1597169.The_End_of_the_World
16. https://philpapers.org/rec/LESTEO-3
17. https://www.rand.org/pubs/commentary/2025/05/could-ai-really-kill-off-humans.html