Citation and evidence

Longtermism

18 min full readUpdated 27 references

This article's verification

Report a problem with this article

Edit article
More

Use this article

Raw MarkdownExplore connections

Improve this page

Revision historyDiscuss this article

Topics

AI EthicsAI Safety

Cite this article

Longtermism is the ethical view that positively influencing the long-term future is a key moral priority of our time.[1] The Oxford philosopher William MacAskill proposed the term in October 2017.[1] Major statements of the view include Toby Ord's book The Precipice (2020), the paper "The Case for Strong Longtermism" by Hilary Greaves and MacAskill, and MacAskill's book What We Owe the Future (2022).[8][2][10] Longtermism grew out of the effective altruism community and was studied at two Oxford research centres, the Future of Humanity Institute and the Global Priorities Institute, both of which have since closed.[11][21][22]

The core argument is that the future could contain an enormous number of people, so actions that change the long-run course of civilisation, above all actions that reduce existential risks, may do far more good in expectation than actions whose benefits are confined to the present.[2][7] Risks from advanced artificial intelligence have been central to that agenda: Greaves and MacAskill used the development of superintelligence as one of their two main examples, and Ord put the chance of an AI-driven existential catastrophe this century at about 10%.[2][9] Longtermism also shaped philanthropy. In 2021 Open Philanthropy organised its grantmaking into a "Longtermism" portfolio and a global health portfolio, and in 2022 the FTX Future Fund, financed mainly by Sam Bankman-Fried, committed more than $130 million, mostly to longtermist causes, before FTX collapsed.[13][14][16] Critics have described it as an ideology that discounts present-day harms in favour of speculative futures, and philosophers have challenged its arguments on decision-theoretic and epistemic grounds.[24][11][25][26]

Definitions and variants

MacAskill introduced the word with a strong definition: "the most important determinant of the value of our actions today is how those actions affect the very long-run future."[1] In a July 2019 post he proposed instead that "longtermism" name a weaker, minimal claim and that the stronger thesis be called "strong longtermism". His minimal definition had three parts: "(i) Those who live at future times matter just as much, morally, as those who live today; (ii) Society currently privileges those who live today above those who will live in the future; and (iii) We should take action to rectify that, and help ensure the long-run future goes well."[1] He also argued for the unhyphenated spelling over "long-termism", and noted that before the term existed the community had relied on clumsy phrases such as "people interested in x-risk reduction".[1]

A November 2021 edit to the same post gave an informal definition, close to the wording later used by the 2025 academic volume Essays on Longtermism: "Longtermism is the view that positively influencing the longterm future is a key moral priority of our time." Strong longtermism, on that wording, holds that it is the key moral priority.[1][3] Ord's own definition calls longtermism "an ethical view that is especially concerned with the impacts of our actions on the longterm future."[5]

VariantCore claimSource
Longtermism (informal)Positively influencing the long-term future is a key moral priority of our timeMacAskill 2019, edited 2021[1]
Strong longtermismIt is the key moral priority; long-run effects are the primary determinant of the value of present actionsMacAskill 2019, edited 2021[1]
Axiological strong longtermismFar-future effects are the most important determinant of the value of our optionsGreaves and MacAskill[2]
Deontic strong longtermismFar-future effects are the most important determinant of what we ought to doGreaves and MacAskill[2]
Minimal longtermismA few far-future risks make mitigation very valuable in a small set of choice situationsGreaves and Tarsney 2025[27]
Expansive longtermismMany ways of improving the far future shape the value of options in all or nearly all choice situationsGreaves and Tarsney 2025[27]

Expanded article table

Greaves and Christian Tarsney argued in 2025 that the standard arguments establish only the minimal form, and that whether the expansive form is more plausible is an open question.[27]

"What we owe the future | Will MacAskill", video from Global Priorities Institute - University of Oxford, published June 5, 2020. Watch on YouTube.

Intellectual background

An early statement of the underlying intuition appears in the philosopher Derek Parfit's Reasons and Persons (1984). Parfit compared three outcomes: peace, a nuclear war that kills 99% of the world's population, and a nuclear war that kills 100%. He held that the difference between the second and third outcomes is much greater than the difference between the first and second, because human extinction would also destroy all of the history that could still follow; Nick Bostrom quoted the passage in his 2013 paper on existential risk.[7]

Nick Bostrom gave the argument a quantitative form. His 2003 paper "Astronomical Waste", published in Utilitas, argued that a very large population could eventually live in the accessible universe, so every year of delay in reaching that future carries an enormous opportunity cost. Bostrom concluded that the lesson for utilitarians was not to speed up technological development but to maximise its safety, meaning the probability that colonisation eventually happens.[6] His 2013 paper "Existential Risk Prevention as Global Priority", in Global Policy, estimated that even if humanity stayed on Earth there could be at least $10^{16}$ future human lives, and inferred that cutting existential risk by one millionth of one percentage point would be worth at least a hundred times the value of a million lives. He proposed a rule of thumb he called "maxipok": "Maximise the probability of an 'OK outcome', where an OK outcome is any outcome that avoids existential catastrophe."[7]

Nick Beckstead's 2013 Rutgers PhD thesis, On the Overwhelming Importance of Shaping the Far Future, argued from broader premises that what matters most is doing what is best for humanity's long-term trajectory. Greaves and MacAskill cite Bostrom and Beckstead as precedents for their own argument.[2]

The case for strong longtermism

Greaves and MacAskill's paper "The Case for Strong Longtermism" first circulated as a Global Priorities Institute working paper in 2019; the revised version appeared as GPI Working Paper No. 5-2021 in June 2021, and a later version was published as a chapter of the 2025 Oxford University Press volume Essays on Longtermism.[11][2][4] Its abstract describes strong longtermism as "the thesis that impact on the far future is the most important feature of our actions today."[4]

The paper defends axiological strong longtermism, stated as a claim about "the most important decision situations facing agents today": that every option that is near-best overall is near-best for the far future, and that every such option delivers much larger benefits in the far future than in the near future. "Far future" means everything after some time $t$, which the authors suggest could be about 100 years after the decision.[2] As a benchmark for the best near-term spending, they used the distribution of insecticide-treated bednets, which they put at about $4,000 per life saved, or 0.025 lives per $100.[2]

The argument then rests on the expected size of the future. Drawing on species lifespans, the habitability of Earth, and the possibilities of space settlement and digital sentience, the authors adopted three estimates of the expected number of future beings:[2]

EstimateExpected number of future beingsWhat it assumes
Main$10^{24}$Any reasonable estimate, in the authors' view, is at least this large
Low$10^{18}$A tiny credence in biological settlement of the Solar System, none in digital sentience or wider settlement
Restricted$10^{14}$Earth-bound life only

Expanded article table

On the restricted estimate, reducing near-term extinction risk by one millionth of one percentage point is worth as much as saving a million lives; on the main estimate it is worth $10^{16}$ lives.[2] To answer the objection that long-run effects "wash out" over time, the authors argued that some actions change the probability that the world ends up in one "persistent state" rather than another. Extinction is the clearest example. Another is a future controlled by artificial superintelligence, whose values could persist as long as civilisation does.[2]

The paper discusses several objections. It argues that the conclusion holds up under plausible departures from expected utility theory, across a wide range of views in population ethics, and whether or not the continuation of civilisation would on balance be good, and it addresses "cluelessness" about long-run effects. On fanaticism, the worry that the case depends on tiny probabilities of enormous payoffs, the authors wrote: "We regard this as one of the most plausible ways in which the argument for strong longtermism might fail." For the deontic version of the thesis, they argued that the decisions they consider, such as where to direct charitable spending or which career to choose, do not require anyone to violate serious moral side-constraints.[2]

Two books presented these ideas to general readers. Toby Ord's The Precipice: Existential Risk and the Future of Humanity was published in March 2020.[8] Ord defines an existential risk as "a risk that threatens the destruction of humanity's longterm potential", and his best guess is that humanity faces "a one-in-six chance of existential catastrophe over the next century".[5] In an interview about the book he put the chance that humanity loses control of the future to unaligned AI this century at about 10%.[9]

MacAskill's What We Owe the Future was published by Basic Books on 16 August 2022. The publisher describes it as an argument for longtermism, "that positively influencing the distant future is our time's key moral priority", and says the book calls on readers to "prepare for a planet where the smartest beings are digital."[10] Time reported at the time of publication that the movement's emphasis on neglected causes had led many effective altruism leaders to focus more on future generations.[13] Elon Musk praised the book as "a close match for my philosophy", a remark later cited by critics including Alexander Zaitchik and Gebru and Torres.[12][11]

Longtermism and AI

Advanced AI, and existential risk from AI in particular, is a recurring focus of longtermist writing. Greaves and MacAskill described two long-run worries about artificial superintelligence: takeover scenarios, in which a misaligned system seizes control, and entrenchment scenarios, in which an authoritarian regime that develops superintelligence first locks in its values indefinitely. They noted that expert judgment tends to put the probability of existential catastrophe from superintelligence at 1 to 10%, and judged that even a highly conservative estimate would be at least 0.1% this century. On their assumptions, $1 billion of well-targeted spending would cut existential risk by at least 0.001%, so every $100 would on average be as valuable as saving one trillion lives on the main estimate and 100 lives on the restricted one, far more than bednets.[2] Their estimate of the size of the future also counts conscious AI systems as potential moral patients.[2]

The paper identified two strands of work on these risks: AI safety research, which aims to make AI systems do what their designers intend, and policy and AI governance work to encourage a cooperative approach between countries and companies, by groups such as the Centre for the Governance of AI. When Greaves and MacAskill wrote, they described both areas as extremely neglected, noting that Open Philanthropy was the only major foundation with these issues as a key focus and spent under $30 million a year on them.[2] Open Philanthropy's longtermist grantmaking was later defined to include its "Potential Risks from Advanced Artificial Intelligence" focus area, alongside biosecurity and effective altruism community growth.[18]

Funding and institutions

Several institutions turned longtermist ideas into research programmes and grants.

InstitutionRoleDates
Future of Humanity Institute (Oxford)Research centre that credits itself with helping originate ideas including existential risk and longtermism2005 to 16 April 2024[22]
Global Priorities Institute (Oxford)Research centre whose publications included work on longtermism, existential risk and population axiology2018 to end of July 2025[21]
Open Philanthropy (Coefficient Giving from November 2025)Grantmaker with a "Longtermism" portfolio from 2021Portfolio named in 2021[14][20]
FTX Future FundGrantmaker for "humanity's long-term prospects"2022 until FTX's collapse in November 2022[15][17]

Expanded article table

In 2021 Open Philanthropy began describing its work as two portfolios. "Global Health and Wellbeing" covered work on health and wellbeing worldwide, while "Longtermism" described "our work to raise the probability of a very long-lasting and positive future"; co-CEO Holden Karnofsky oversaw the longtermist grantmaking.[14] Time reported that of the more than $440 million Open Philanthropy granted in 2021, 28% went to longtermist interventions such as biosecurity and effective altruism community growth.[13] In November 2025 the organisation renamed itself Coefficient Giving and said its former focus areas were becoming funds, among them a Navigating Transformative AI Fund and a Global Catastrophic Risks Opportunities Fund.[20]

The FTX Foundation's Future Fund described itself in 2022 as supporting "ambitious projects to improve humanity's long-term prospects", with interests including "the safe development of artificial intelligence, reducing catastrophic biorisk, improving institutions, effective altruism, and more."[15] It was funded primarily by Bankman-Fried, with major contributions from Caroline Ellison, Gary Wang and Nishad Singh. Nick Beckstead served as chief executive of the FTX Foundation, and MacAskill was a member of the fund's team.[16] By August 2022 it had committed more than $130 million in grants.[13]

When FTX collapsed in November 2022, the Future Fund team resigned, writing that they had "fundamental questions about the legitimacy and integrity of the business operations" and could no longer process grants.[17] Open Philanthropy announced that it was "pausing most new longtermist funding commitments" in AI risk, biosecurity and effective altruism community growth while it decided how far to raise its funding bar.[18] It lifted the pause in January 2023; Karnofsky estimated that about 40 to 70% of the grants made over the previous 18 months would have cleared the new, higher bar.[19]

Criticism

Political and ideological critiques

In an October 2021 essay for Aeon, "Against longtermism", Émile P. Torres argued that the view makes catastrophes short of extinction look minor next to humanity's cosmic potential, that its expected-value arithmetic could be used to justify extreme measures, and that its utilitarian framing treats people as containers of value. The essay quotes Bostrom's line that "even the worst of these catastrophes are mere ripples on the surface of the great sea of life."[24]

In a 2024 paper in First Monday, Torres and the computer scientist Timnit Gebru grouped longtermism with six other ideologies under the acronym TESCREAL: "transhumanism, Extropianism, singularitarianism, (modern) cosmism, Rationalism, Effective Altruism, and longtermism". They argued that this bundle traces back to twentieth-century eugenics and drives the race to build artificial general intelligence in ways that harm marginalised groups and centralise power.[11]

Other writers have raised related objections. Writing in The New Republic in October 2022, Alexander Zaitchik argued that longtermism asks people to sacrifice present well-being for hypothetical future value and that its real stakes lie in the billionaire politics it promotes, citing Musk's endorsement.[12] Even before the FTX collapse, Time reported that some people inside effective altruism worried that the shift of money and attention toward existential risk could sideline the movement's original focus on pressing, neglected suffering.[13]

Philosophical critiques

Greaves and MacAskill themselves identified fanaticism, the reliance on tiny probabilities of enormous payoffs, as one of the most plausible ways their argument could fail.[2]

David Thorstad, writing in Philosophy & Public Affairs in 2023, addressed two claims that many philosophers defend together: that mitigating existential risk is astronomically valuable, and that humanity faces high levels of existential risk. He argued that the second claim significantly reduces the value of mitigation, enough to threaten the first, and drew from this reduced prospects for ethical longtermism.[25] Eric Schwitzgebel, in a 2026 Utilitas article, argued that no one can justifiably believe any currently available action will have a non-negligible positive influence a billion or more years from now. His three arguments are that finite effects wash out once infinite effects are possible, that we cannot tell which actions will help on that timescale, and that compounding probabilities make any such effect negligible.[26]

Later developments

The 2025 volume Essays on Longtermism, edited by Greaves, Jacob Barrett and David Thorstad and published by Oxford University Press under a Creative Commons licence, collects contributions from supporters and critics drawn from philosophy, economics, psychology, law, political science, mathematics and private industry. It is organised around four questions: whether to accept some version of longtermism, how far the far future can be predicted and controlled, which priorities longtermism recommends, and what it implies for social and political institutions.[3]

The Global Priorities Institute closed at the end of July 2025, saying its research threads would continue through a wider community of researchers.[21] In August 2025 MacAskill published "Better Futures", an essay series from the AI macrostrategy research group Forethought, arguing that improving how good the future is, in scenarios where humanity survives, deserves roughly the same priority as raising the chance of survival.[23]

Timeline

DateEvent
1984Parfit's three-outcomes comparison appears in Reasons and Persons[7]
November 2003Bostrom publishes "Astronomical Waste" in Utilitas[6]
February 2013Bostrom publishes "Existential Risk Prevention as Global Priority"[7]
2013Beckstead completes his thesis on shaping the far future[2]
October 2017MacAskill proposes the term "longtermism"[1]
2018Global Priorities Institute founded at Oxford[21]
July 2019MacAskill publishes his definitions of longtermism and strong longtermism[1]
March 2020Ord's The Precipice published[8]
June 2021"The Case for Strong Longtermism" issued as GPI Working Paper No. 5-2021[2]
2021Open Philanthropy names its "Longtermism" portfolio[14]
October 2021Torres publishes "Against longtermism" in Aeon[24]
August 2022What We Owe the Future published[10]
November 2022FTX collapses; Future Fund team resigns; Open Philanthropy pauses new longtermist commitments[17][18]
January 2023Open Philanthropy lifts the pause[19]
April 2024Future of Humanity Institute closes; Gebru and Torres publish the TESCREAL paper[22][11]
July 2025Global Priorities Institute closes[21]
August 2025Essays on Longtermism published; MacAskill publishes "Better Futures"[3][23]
November 2025Open Philanthropy becomes Coefficient Giving[20]
June 2026Schwitzgebel's "The Washout Argument against Longtermism" published online[26]

Expanded article table

Continue reading

More from See also

More to read

Topics:AI EthicsAI Safety

References

  1. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8 ^9 ^10William MacAskill. "'Longtermism'". Effective Altruism Forum, 25 July 2019 (edited November 2021). forum.effectivealtruism.org/...longtermism
  2. ^1 ^2 ^3 ^4 ^5 ^6 ^7 ^8 ^9 ^10 ^11 ^12 ^13 ^14 ^15 ^16 ^17 ^18 ^19Hilary Greaves and William MacAskill. "The Case for Strong Longtermism". GPI Working Paper No. 5-2021, Global Priorities Institute, June 2021. globalprioritiesinstitute.org/...June-2021-2-2.pdf
  3. ^1 ^2 ^3Hilary Greaves, Jacob Barrett and David Thorstad (eds.). "Essays on Longtermism: Present Action for the Distant Future". Oxford University Press, August 2025. doi.org/...9780191979972.001.0001
  4. ^1 ^2Hilary Greaves and William MacAskill. "The Case for Strong Longtermism". In Essays on Longtermism, Oxford University Press, 2025, pp. 17-49. doi.org/...9780191979972.003.0003
  5. ^1 ^2Toby Ord. "Frequently Asked Questions". The Precipice (book website). theprecipice.com/faq
  6. ^1 ^2Nick Bostrom. "Astronomical Waste: The Opportunity Cost of Delayed Technological Development". Utilitas 15(3): 308-314, November 2003. doi.org/...s0953820800004076
  7. ^1 ^2 ^3 ^4 ^5Nick Bostrom. "Existential Risk Prevention as Global Priority". Global Policy 4(1): 15-31, February 2013. doi.org/...1758-5899.12002
  8. ^1 ^2 ^3Hachette Book Group. "The Precipice: Existential Risk and the Future of Humanity" by Toby Ord. Grand Central Publishing, 24 March 2020. hachettebookgroup.com/...9780316484893
  9. ^1 ^2Robert Wiblin, Arden Koehler and Keiran Harris. "#72 - Toby Ord on the precipice and humanity's potential futures". The 80,000 Hours Podcast, 7 March 2020. 80000hours.org/...existential-risk-future-humanity
  10. ^1 ^2 ^3Hachette Book Group. "What We Owe the Future" by William MacAskill. Basic Books, 16 August 2022. hachettebookgroup.com/...9781541618633
  11. ^1 ^2 ^3 ^4 ^5 ^6Timnit Gebru and Émile P. Torres. "The TESCREAL bundle: Eugenics and the promise of utopia through artificial general intelligence". First Monday 29(4), 14 April 2024. doi.org/...fm.v29i4.13636
  12. ^1 ^2Alexander Zaitchik. "The Heavy Price of Longtermism". The New Republic, 24 October 2022. newrepublic.com/...ture-humanity-william-macaskill
  13. ^1 ^2 ^3 ^4 ^5Naina Bajekal. "Want to Do More Good? This Movement Might Have the Answer". Time, 10 August 2022. time.com/...ongtermism-william-macaskill-interview
  14. ^1 ^2 ^3 ^4Open Philanthropy. "Our Progress in 2021 and Plans for 2022". Open Philanthropy, 2022 (archived 6 June 2022). web.archive.org/...our-progress-2021-and-plans-2022
  15. ^1 ^2FTX Foundation. "Future Fund: Let's build a flourishing future" (homepage). Future Fund, archived 9 August 2022. web.archive.org/...ftxfuturefund.org
  16. ^1 ^2FTX Foundation. "About" (Future Fund). Future Fund, archived 18 September 2022. web.archive.org/...about
  17. ^1 ^2 ^3Nick Beckstead, Leopold Aschenbrenner, Avital Balwit, Ketan Ramakrishnan and Will MacAskill. "The FTX Future Fund team has resigned". Effective Altruism Forum, 11 November 2022. forum.effectivealtruism.org/...team-has-resigned-1
  18. ^1 ^2 ^3Holden Karnofsky. "Some comments on recent FTX-related events". Effective Altruism Forum, 10 November 2022. forum.effectivealtruism.org/...-ftx-related-events
  19. ^1 ^2Holden Karnofsky. "We're no longer 'pausing most new longtermist funding commitments'". Effective Altruism Forum, 30 January 2023. forum.effectivealtruism.org/...longtermist-funding
  20. ^1 ^2 ^3Aaron Gertler. "Open Philanthropy Is Now Coefficient Giving". Effective Altruism Forum, 18 November 2025. forum.effectivealtruism.org/...-coefficient-giving
  21. ^1 ^2 ^3 ^4 ^5Global Priorities Institute. "Global Priorities Institute" (homepage). University of Oxford, accessed November 2026. globalprioritiesinstitute.org
  22. ^1 ^2 ^3Future of Humanity Institute. "Future of Humanity Institute" (archive homepage). Accessed November 2026. futureofhumanityinstitute.org
  23. ^1 ^2William MacAskill. "Better Futures". Forethought, 3 August 2025. forethought.org/...better-futures
  24. ^1 ^2 ^3Émile P. Torres. "Against longtermism". Aeon, 19 October 2021. aeon.co/...the-worlds-most-dangerous-secular-credo
  25. ^1 ^2David Thorstad. "High Risk, Low Reward: A Challenge to the Astronomical Value of Existential Risk Mitigation". Philosophy & Public Affairs 51(4): 373-412, 2023. doi.org/...papa.12248
  26. ^1 ^2 ^3Eric Schwitzgebel. "The Washout Argument against Longtermism". Utilitas, published online 16 June 2026. doi.org/...s095382082610034x
  27. ^1 ^2 ^3Hilary Greaves and Christian Tarsney. "Minimal and Expansive Longtermism". In Essays on Longtermism, Oxford University Press, 2025, pp. 315-333. doi.org/...9780191979972.003.0021

Improve this article

Add missing citations, update stale details, or write a clearer explanation. Anyone can edit: changes go live when you publish, and every edit is checked by an editor.

2 revisions · v2 · 3,615 words · full history

Fact-checks are independent of edits: a reviewer re-verifies the article against its sources and stamps the date. How we verify

Research and drafting on this wiki are AI-assisted, under named human editorial standards. How AI is used here

Reviewer note: Independent fact-check before publication on November 18, 2026: 96 claims checked against 31 cited and independent sources; 1 material and 6 minor corrections applied before publication.

Cite this page: AI Wiki. "Longtermism." aiwiki.ai, updated 18 Nov 2026, fact-checked 18 Nov 2026. CC BY 4.0. https://aiwiki.ai/wiki/longtermism

Edit article