Episode Summary
Executive Summary: Will MacAskill traces effective altruism’s origins and evolution, then argues for taking moral uncertainty seriously via expected-value-style reasoning, while warning about infinite-value and inter-theoretic comparison problems. He defends utilitarianism more strongly than common sense, links EA to longtermism and moral reflection, and critiques both EA’s public messaging mistakes and academia’s incentives.
Main Topics: Origins and growth of effective altruism: MacAskill explains EA as the convergence of GiveWell, Les Long, and Giving What We Can, then describes its expansion into a global community focused on doing the most good across causes. Moral uncertainty and decision theory: A major portion of the interview develops his book project on moral uncertainty, arguing that agents should not simply act as if one moral theory is certainly correct, but should weigh theories probabilistically while grappling with hard comparison problems. Inter-theoretic comparison, structuralism, and fanaticism: He discusses why comparing moral theories is hard, including theories with infinite value, incomparability, and the challenge of assigning cardinal magnitudes across moral views; he favors allowing amplified theories over flattening all theories via normalization. Arguments for classical utilitarianism: MacAskill presents several defenses of utilitarianism: its historical track record, Harsanyi-style impartiality, and Parfit-inspired skepticism about personhood as a deep moral boundary. Practical implications for EA and longtermism: He argues for a hybrid of consequentialism plus rights, strong concern for the long-run future, animals, and moral information, and emphasizes option value and the long reflection. Academia, publishing, and career advice: MacAskill reflects on the downsides of academic life, perverse publishing incentives, the importance of networks, and why philosophy may still be worth studying despite poor job prospects, though economics may be a better alternative for many. EA’s messaging, diversity, and mistakes: He criticizes early EA marketing around earning to give, says the movement is often misrepresented in media, and urges greater epistemic and demographic diversity plus more careful communication around controversial ideas.
Key Arguments: Effective altruism began as a confluence of three movements—GiveWell, Les Long, and Giving What We Can—and grew because the core question 'how do we do the most good?' resonated globally. Moral uncertainty should be handled more like empirical uncertainty: assign probabilities to moral theories and act on expected choiceworthiness rather than merely picking a favorite theory. Structuralist attempts to normalize moral theories can be too restrictive; amplified theories may preserve real differences in value structure, including cases where one theory treats the same rank ordering as far more important than another. Fanaticism and infinite-value theories create real problems for expected-value calculations, but those problems do not show such theories are false; they show the need for better decision rules or a broader epistemic posture. Utilitarianism is defended by its historical success in anticipating moral progress (e.g., slavery, women’s rights, animal welfare, sexual liberty) and by impartiality arguments like Harsanyi’s veil of ignorance. Rejecting personal identity as morally fundamental supports utilitarian trade-offs across persons, since the same reasoning used to trade off pain and pleasure within a life may apply across lives. EA should combine consequentialist ambition with side constraints/rights, especially against direct rights violations, while remaining strongly open to longtermist causes, animals, and future generations. Because moral knowledge is fragile and history is full of confident moral error, preserving option value and enabling future moral reflection are major priorities; this supports reducing existential risk and pursuing a 'long reflection.' Academia is valuable for foundational theory but often inefficient, bureaucratic, and distorted by publication and prestige incentives; applied impact may be better pursued outside academia in think tanks or EA organizations. EA’s biggest self-inflicted mistake was early aggressive marketing around earning to give, which left a long-lived public misperception that the movement still struggles to undo.
Data Points: co-founded organizations: 3 - MacAskill co-founded Giving What We Can, 80,000 Hours, and the Center for Effective Altruism. local EA groups: about 100 - He says there are now about 100 local groups dedicated to EA ideas. EA community size: thousands - He describes a 'vibrant community of thousands' worldwide. CEA CEO tenure returned: about 1.5 years - He says he returned as CEO of CEA about a year and a half before stepping away. Oxford hiring committee weight for test vs interview: 8 to 1 - MacAskill says Oxford data suggested tests should count about eight times as much as interviews. EA core agreement on longtermism: 80-90% - He says a recent survey of core EA members found around 80–90% agreement with longtermism. philosophers' views on animals: two-thirds - He notes about two-thirds of philosophers believe it is obligatory not to eat meat. moral-philosophy survey distribution: roughly one-third each - He references a survey with roughly one-third consequentialists, one-third deontologists, and one-third virtue ethicists/other. publication lag: 5 years - He says the gap between submitting and publishing his second major publication was five years. book timeline: 1 to 1.5 years - He estimates his moral uncertainty book will come out in about a year to a year and a half.
Pivotal Quotes: "How can I do as much good with at least a significant part of my resources?" — Will MacAskill: He summarizes the core motivating question behind effective altruism. "The football fan model of decision-making given moral uncertainty" — Will MacAskill: His label for people who act as if one moral theory is obviously correct instead of reasoning under uncertainty. "I think the biggest owned goal in the history of moral philosophy" — Will MacAskill: His criticism of Rawls' veil-of-ignorance argument as he presents it.
Implications: Listeners should expect EA to keep shifting toward longtermism, moral uncertainty, and broader cause prioritization. For researchers and careers, the episode argues for foundational work on ethics, population, and existential risk, plus more careful public communication and institutional reform.