# Longtermism The philosophical position that the primary moral obligation of the present is to ensure the long-term flourishing of future humanity — potentially trillions of people spread across billions of years and, eventually, the stars. Developed primarily by philosophers at the University of Oxford, longtermism is the "L" in [[TESCREAL]] and arguably the most politically consequential component of the bundle: it provides the moral framework that converts abstract futurism into concrete policy, funding decisions, and power concentration. **Core thesis for this KB:** Longtermism matters not because its questions are trivial — existential risk is real — but because its framework systematically devalues present-day suffering, concentrates decision-making power in a self-selected techno-elite, and provides intellectual cover for policies (AI deregulation, social program defunding, democratic skepticism) that happen to align perfectly with the material interests of its richest adherents. → Part of the [[TESCREAL]] bundle. See also: [[Effective Altruism]], [[Russian Cosmism]], [[Transhumanist Defense]] --- ## Origins and Key Architects ### [[Nick Bostrom]] and the [[Existential Risk]] Framework Swedish-born philosopher Nick Bostrom is the intellectual architect of longtermism, though the term itself came later. At the University of Oxford, Bostrom founded the [[Future of Humanity Institute]] (FHI) in 2005 and spent two decades developing the conceptual infrastructure: existential risk as a philosophical category, the idea that "astronomical" numbers of future people should weigh heavily in moral calculations, and the argument that preventing human extinction is the single most important thing anyone could work on. His 2014 book *[[Superintelligence (Book)]]* became the canonical text for AI existential risk — arguing that artificial general intelligence, if misaligned with human values, could pose an extinction-level threat. The book was enormously influential in Silicon Valley: [[Elon Musk]] publicly cited it, [[Bill Gates]] endorsed it, and it catalyzed a wave of AI safety funding and research. **FHI closure:** In April 2024, the University of Oxford closed the Future of Humanity Institute after years of administrative conflict. Oxford's Faculty of Philosophy imposed a fundraising and hiring freeze starting in 2020, leading to a gradual loss of researchers. Bostrom described it as "death by bureaucracy." He subsequently resigned from Oxford and founded the Macrostrategy Research Initiative. The closure of longtermism's flagship institution came amid the broader post-FTX credibility crisis engulfing the movement. **The racist email:** In January 2023, Bostrom preemptively published a 1996 email in which he used explicitly racist language, including the n-word, and speculated about racial differences in intelligence. His apology was widely criticized as inadequate — he expressed regret for the language while defending the underlying intellectual inquiry as legitimate. The incident confirmed what critics like [[Émile Torres]] had been arguing: that longtermism's intellectual lineage runs through eugenics, scientific racism, and colonial "civilizing mission" ideology. ### [[William MacAskill]] and the Popularization Scottish philosopher William MacAskill co-developed longtermism at Oxford and became its most prominent public advocate. His 2022 book *What We Owe the Future* brought longtermism to a mass audience, arguing that future people are moral patients whose interests should weigh in our decisions today, and that we live at a uniquely pivotal moment in human history. MacAskill is also the co-founder of the [[Effective Altruism]] movement, and the longtermist turn within EA was largely his doing. He argued that EA's original focus on tractable near-term problems (malaria prevention, deworming) was thinking too small — that the *really* high-impact interventions involved shaping the long-term trajectory of civilization, particularly by mitigating AI risk. **The SBF connection:** MacAskill was [[Sam Bankman-Fried]]'s philosophical mentor. He introduced SBF to EA ideas and encouraged him to pursue earning to give — making as much money as possible in order to donate it to EA-aligned causes. When FTX collapsed in November 2022, revealing a massive fraud, the moral framework that had justified SBF's risk-taking (longtermist logic: if the expected value of your donations is high enough, extreme financial risk is justified) came under devastating scrutiny. MacAskill's public statement expressed shock, but critics noted that EA leadership had been warned about SBF's unethical behavior as early as 2018. ### [[Toby Ord]] and *The Precipice* Oxford philosopher Toby Ord's 2020 book *The Precipice: Existential Risk and the Future of Humanity* formalized the existential risk framework. Ord estimated a roughly 1-in-6 chance of existential catastrophe in the next century — with unaligned AI as the largest risk — and argued that reducing this risk should be civilization's top priority. The book became required reading in EA and longtermist circles and heavily influenced the AI safety policy discourse. --- ## The Core Arguments ### The Astronomical Stakes Argument The mathematical heart of longtermism: if humanity survives and eventually colonizes space, the number of future people could be astronomically large — trillions upon trillions over billions of years. Therefore, any action that even slightly increases the probability of long-term human flourishing has an expected value that dwarfs anything we could do for people alive today. In their paper "The Case for Strong Longtermism," Hilary Greaves and MacAskill demonstrate this with a provocative calculation: assigning a mere 0.001% probability to AI-driven catastrophe and running the expected value math produces results where $100 spent on AI safety saves more expected lives than thousands spent on malaria nets. The appeal is obvious. So is the problem: when your moral framework says that speculative future benefits always outweigh concrete present suffering, you've built a machine for rationalizing neglect. ### The Pivotal Moment Argument Longtermists argue we are living through a uniquely consequential period — a "hinge of history" — because transformative technologies (AI, biotech, nanotech) could lock in trajectories that determine the long-run future. If AI development goes well, utopia; if it goes wrong, extinction or permanent dystopia. Therefore, influencing AI development *right now* is the most important thing anyone can do. This argument does double duty: it creates urgency (we can't wait for democratic deliberation) and it elevates the people who happen to work on AI (they're not just engineers — they're saving the future). ### The Elite Stewardship Argument The less-stated but logically necessary conclusion: if the stakes are this high and the window this narrow, then the people best positioned to navigate it — AI researchers, longtermist philosophers, and their funders — should have outsized influence over decisions that affect everyone. Democratic processes are too slow, too uninformed, and too captured by short-term thinking. This is where longtermism converges with the [[Dark Enlightenment]]'s anti-democratic politics, despite radically different starting premises. Both conclude that mass democracy is inadequate to the challenges ahead and that an enlightened elite should steer. --- ## The [[Eugenics]] Connection One of the most explosive aspects of the Torres-Gebru critique is the argument that longtermism is a direct intellectual descendant of the Anglo-American eugenics tradition. ### The Intellectual Genealogy Torres traces a line from Francis Galton (who coined "eugenics") through Julian Huxley (who coined "transhumanism" and led the British Eugenics Society) to the modern TESCREAL bundle. The connecting logic: eugenicists believed they had a duty to improve the future of the species through selective breeding; longtermists believe they have a duty to improve the future of the species through technological intervention. The frame has shifted from genetics to AI, but the underlying structure — a self-appointed elite deciding what's best for the future of humanity — remains intact. ### Bostrom's Track Record Bostrom's 1996 racist email was not an isolated incident in the context of his intellectual output. His work on existential risk includes discussions of "dysgenic" trends (the idea that less intelligent people are outbreeding more intelligent ones), and his 2004 paper "The Future of Human Evolution" discusses scenarios involving genetic selection. Torres argues these aren't aberrations — they're the logical endpoint of a framework that treats future "quality" of humanity as more important than present equality. ### The Pronatalist Thread The modern pronatalist movement — led by figures like Simone and Malcolm Collins, who advocate for IVF and embryo selection to produce "higher-quality" children — is explicitly connected to longtermist ideas and funded by people in the [[Peter Thiel]] orbit. The Collins frame reproduction as a civilizational imperative, echoing longtermist arguments about ensuring the future of humanity. See: [[Transhumanist Defense]] for the defense dimensions of pronatalism. --- ## Longtermism's Political Functions ### Justifying AI Deregulation If AI development determines the fate of civilization, and if the biggest risk is that America *doesn't build fast enough* (the argument in [[Leopold Aschenbrenner]]'s influential 2024 paper *[[Situational Awareness]]*), then regulation becomes an existential threat. Longtermism provides the intellectual framework for the defense tech lobby's argument that AI safety means AI supremacy, not AI caution. ### Defunding Present-Day Programs If future people matter more than present people (the strong longtermist claim), then redirecting resources from social programs, climate adaptation, public health, and poverty reduction toward AI research, space colonization, and biosecurity is not just permissible but morally obligatory. This conclusion happens to align perfectly with the libertarian policy preferences of longtermism's wealthiest backers. ### Concentrating AI Governance Longtermist AI safety discourse often concludes that AI development should be concentrated in a small number of well-resourced labs that can afford to do safety research — and that open-source AI is dangerous because it distributes power too widely. This framework serves the competitive interests of labs like [[OpenAI]], [[Anthropic]], and DeepMind while wearing the mask of existential risk reduction. --- ## The Institutional Infrastructure ### Academic Homes - **Future of Humanity Institute (FHI)** — Oxford. Founded 2005, closed April 2024. Longtermism's flagship for two decades. - **[[Global Priorities Institute]] (GPI)** — Oxford. Founded 2018 by Hilary Greaves. Academic research on longtermist priorities. - **[[Centre for the Study of Existential Risk]] (CSER)** — Cambridge. Founded 2012 by Martin Rees, [[Jaan Tallinn]], and Huw Price. ### Funding Networks - **[[Open Philanthropy]] / Coefficient Giving** (renamed November 2025) — [[Dustin Moskovitz]]'s mega-foundation. Largest funder of longtermist-aligned research, spending ~$46 million on AI safety in 2023 alone. Over $4 billion total directed across focus areas by mid-2025. - **[[Survival and Flourishing Fund]]** — Jaan Tallinn (Skype co-founder). Major funder of longtermist and rationalist organizations. - **FTX Future Fund** — SBF's longtermist philanthropy vehicle. Shut down in November 2022 after FTX collapse. ### Movement Infrastructure - **[[80,000 Hours]]** — EA career advice organization that funnels talent toward longtermist-aligned work - **Effective Altruism Forum** — Online hub for EA/longtermist discourse - **EA Global conferences** — Networking events connecting academics, funders, and operators --- ## The Critiques ### Émile Torres: "The World's Most Dangerous Secular Credo" Torres's critique is the most comprehensive. Writing in Aeon, Torres argues that longtermism "might be the most dangerous secular belief system in the world" — not because existential risk isn't real, but because longtermism's framework: 1. Systematically devalues the lives and suffering of people alive today 2. Provides moral cover for ignoring climate change (which longtermists typically don't classify as an existential risk) 3. Concentrates power in a small, unaccountable elite 4. Has direct intellectual roots in eugenics and scientific racism 5. Generates conclusions (like SBF's "earning to give" at any cost) that justify harmful actions through speculative future benefits ### Alice Crary: "Toxic Ideology" Writing in *Radical Philosophy*, philosopher Alice Crary argues longtermism isn't just wrong but actively harmful — functioning as a "galaxy-brained" rationalization for elite indifference to present injustice. ### The Climate Blind Spot Longtermists typically classify climate change as a serious problem but not an existential risk — meaning it gets deprioritized relative to AI risk and engineered pandemics. Critics at the *Bulletin of the Atomic Scientists* and elsewhere argue this reflects both a miscalculation of climate tail risks and a convenient alignment with the fossil fuel interests of longtermism's tech-libertarian funders. --- ## Connections to the Right-Wing KB **The TESCREAL pipeline:** Longtermism is the moral engine of [[TESCREAL]]. It converts the bundle's techno-futurism into actionable priorities and provides the ethical framework for concentrating resources in AI development, defense tech, and elite governance. See: [[TESCREAL#The TESCREAL-to-Power Pipeline]]. **Defense tech endpoint:** The journey from "AI poses existential risk" to "therefore America needs AI weapons supremacy" is the longtermism-to-defense pipeline. See: [[Transhumanist Defense#The TESCREAL-to-Defense Pipeline]]. **The Unholy Alliance:** Longtermism functions as the secular eschatology of the tech wing of [[The Unholy Alliance]] — parallel to Christian nationalism's religious eschatology, but arriving at compatible political conclusions (elite governance, democratic skepticism, civilizational urgency). **Dark Enlightenment convergence:** Longtermism and [[Dark Enlightenment]] reach the same anti-democratic conclusions through different reasoning. NRx argues democracy is inherently dysfunctional; longtermism argues democracy is too slow for existential stakes. The practical policy outcomes overlap significantly. --- ## See Also - [[TESCREAL]] — The bundle framework; longtermism is the "L" - [[Effective Altruism]] — The movement longtermism captured and redirected - [[Russian Cosmism]] — Longtermism's deepest philosophical ancestor - [[Effective Accelerationism]] — The accelerationist fork that shares longtermist conclusions - [[Transhumanist Defense]] — The defense tech endpoint of the longtermist pipeline - [[Dark Enlightenment]] — Anti-democratic twin ideology - [[MIRI]] — Institutional bridge between rationalism and AI safety/longtermism - [[Mars]] — the "backup planet" colonization narrative and escape-hatch ideology