Effective altruism

Working Definition

Effective altruism (EA) is a philosophical stance and social movement that treats doing good as an optimization problem: because resources are finite and suffering is measurable, we ought to choose the actions that produce the greatest expected improvement in well-being per unit of cost, guided by empirical evidence rather than warm impulse. Its everyday version boils down to “give smart and vet charities.” Its technical version rests on three much stronger premises: impartiality (a stranger’s life counts exactly as much as a neighbor’s), commensurability (diverse kinds of good can be translated into a common metric), and maximization (the gap between a good action and the best available action is morally decisive). Its analytical payoff lies in forcing explicit comparison where traditional philanthropy has long relied on unexamined loyalties.

Genesis and Global Lineage

The philosophical core comes from Peter Singer, who argued in his 1972 article Famine, Affluence, and Morality that geographical distance does not weaken the duty to save a life: just as we are obligated to wade into a shallow pond to rescue a drowning child even at the cost of ruining our clothes, we are obligated to aid people dying of famine in Bengal. For three decades, the essay remained largely a seminar room exercise in utilitarianism. Two groups turned it into an institutional network: the hedge-fund analysts Holden Karnofsky and Elie Hassenfeld, who founded GiveWell in 2007 to evaluate charities the way investors evaluate firms, and the Oxford philosophers Toby Ord and William MacAskill, who launched Giving What We Can in 2009 around a public pledge to donate at least 10% of one’s lifetime income. The movement acquired its name in December 2011, when Giving What We Can and the career-advice organization 80,000 Hours voted to incorporate their shared umbrella body as the Centre for Effective Altruism. Its popular canon comprises Singer’s The Life You Can Save (2009) and The Most Good You Can Do (2015), MacAskill’s Doing Good Better (2015), and Ord’s The Precipice (2020). Over time, the concept was partially captured by two neighboring circuits: the online rationalist community (LessWrong, superforecasting, AI safety), which shifted the movement’s center of gravity from anti-malaria bed nets toward existential risk, and Silicon Valley tech philanthropy, which supplied both billions of dollars and severe reputational liabilities.

Central European Lineage

We found no documented link between this concept and the intellectual history or historical experience of Central Europe.

Why Now

The years between 2018 and 2026 brought effective altruism first to its zenith of influence and then into acute crisis. By 2021, the movement was steering billions of dollars committed by Open Philanthropy (backed by Dustin Moskovitz and Cari Tuna) and by Sam Bankman-Fried, whose FTX Foundation launched in February 2021 as the flagship sponsor of EA’s longtermist wing. In August 2022 MacAskill published What We Owe the Future to wide media fanfare; three months later, in November 2022, the FTX cryptocurrency exchange collapsed, its foundation shut down, and Bankman-Fried was convicted of fraud in November 2023 and sentenced in March 2024 to 25 years in prison. For critics, the episode proved that an ethics of expected-value maximization stripped of constraints against risk and deception predictably breeds exactly such actors. At the same time, EA organizations had become the primary philanthropic pipeline funding technical AI safety research and policy fellowships, putting the movement at the center of post-2023 battles over AI governance, lab boardrooms (most visibly during the November 2023 OpenAI board crisis), and who gets to define “existential” threats.

Explanatory Power

The concept brings into focus three structural facts that conventional charity obscures. First, the heavy-tailed distribution of effectiveness: the difference between the best and the median global-health intervention is not 20%, but often a factor of a hundred, meaning that choosing which program to fund matters far more than how generously one gives; this explains why GiveWell channels hundreds of millions of dollars into a handful of interventions (insecticide-treated malaria nets, seasonal malaria chemoprevention, vitamin A supplementation, direct cash transfers) rather than thousands of local projects. Second, the opportunity cost of a career: the controversial strategy of “earning to give” highlights that a medical student in a wealthy country largely replaces another qualified applicant at the margin, whereas a high-earning professional donating a substantial share of income to vaccinations can save far more lives than practicing medicine directly. Third, the triage heuristic of scale, neglectedness, and tractability for selecting entire cause areas rather than individual NGOs—explaining why the very same movement funds campaigns against factory farming of broiler chickens alongside pandemic pathogen surveillance. Without the concept, that portfolio looks eccentric; through the concept, it is the output of a single decision algorithm.

Axes of Debate

Underappreciated Angle

The most consequential innovation of effective altruism lies not in moral philosophy, but in institutional epistemology. More than any prior philanthropic movement, EA introduced publicly audited cost-effectiveness models, open spreadsheets, explicit uncertainty intervals, and post-mortems of failed grants into a sector that had historically accounted for itself through good intentions and donor prestige. It imported the norms of open science into charity. The paradox is that this very culture of transparent quantification, when paired with a single shared optimization formula, breeds an intellectual monoculture: thousands of donors and foundations funnel capital toward whatever a few centralized evaluators rank highest, while problems that resist formal modeling vanish from view—not because they matter less, but because no one knows how to put an error bar on them. Goodhart’s law operates here at the scale of an entire philanthropic ecosystem: the metric of effectiveness becomes the target, and the methodological pluralism that once buffered philanthropy against systemic blind spots is dismissed as waste. How to preserve empirical rigor without creating an institutional monoculture is a deeper challenge for the movement’s future than the headlines over FTX.