Episode Summary
Executive Summary: Will MacAskill argues for long-termism: the view that we should give far more weight to future people because our actions can shape an enormous future. The conversation explores why this differs from sustainability, why existential risks and value-shaping matter, and where practical leverage lies today—especially pandemic preparedness, AI alignment, forecasting, and clean tech—while also addressing criticism, feedback limitations, and individual actions like giving and career choice.
Main Topics: What long-termism is (Priority: 5/5): MacAskill defines long-termism as the idea that we should do much more to protect future generations because future people matter morally and there may be vastly many of them. Long-termism vs. sustainability (Priority: 4/5): He contrasts long-termism with sustainability, arguing that future people should be considered using the same moral reasoning as people alive today, not via a separate ethical category. Existential risk and future value (Priority: 5/5): The discussion emphasizes preventing human extinction and other catastrophic events, since they could erase all future generations and the enormous potential value they contain. Value shaping and historical contingency (Priority: 5/5): MacAskill argues that today’s values may determine the trajectory of civilization for centuries or longer, so influencing norms and institutions matters, not just avoiding catastrophe. Practical long-termist priorities (Priority: 5/5): He identifies pandemic preparedness, AI alignment, forecasting, and clean technology as especially promising areas because they are both impactful and, in some cases, somewhat measurable in the near term. Effective altruism, criticism, and feedback loops (Priority: 4/5): The interview examines how effective altruism relates to long-termism, the challenge of weak feedback on far-future interventions, and why the movement should remain diverse across causes. Individual action and public reception (Priority: 3/5): MacAskill says individuals can help through donations and career choice, discusses the media response to the book, and addresses misinformation and extreme anti-human views.
Key Arguments: Future generations matter morally, and the sheer number of possible future people makes their interests extremely important. Long-termism applies the same moral reasoning we use for present-day people across countries to people in the far future. Existential risks such as pandemics, nuclear war, and misaligned AI are urgent because they could end civilization or permanently damage its trajectory. Long-term impact is not only about preventing extinction; it also includes shaping the values and institutions that guide future societies. Pandemic preparedness is a high-leverage cause because many risks were foreseeable, and the post-COVID response has been inadequate. Some long-termist work can still have near-term feedback, such as wastewater monitoring and air sterilization technologies, even if the ultimate goal is long-horizon benefit. The effective altruism movement should remain pluralistic, because different people can specialize in different high-impact areas and the field is still learning. Individuals in rich countries can meaningfully help through targeted giving and by choosing careers strategically, not only through activism or policymaking.
Data Points: Future time horizon: 100 years to 100 million years (and in principle indefinitely) - MacAskill explains that long-termism can extend across very long time horizons Humanity’s possible life expectancy: Hundreds of millions or even billions of years - He argues humanity may endure far longer than people typically assume Pandemic prediction (2016): One in three probability - He says the EA community predicted a pandemic with at least 10 million deaths in the next decade Pandemic death estimate: 23 million - He cites current excess mortality estimates from COVID-19 US pandemic funding proposal: $70 billion - He references the Apollo program funding in Build Back Better that stalled in Congress Human challenge trial volunteers: 60,000 - He describes One Day Sooner volunteers willing to accept infection risk to speed vaccine development Vaccine timeline: Months earlier - He says challenge trials could have enabled earlier vaccine rollout Public well-being survey: Over 70% - He says people in the US and India report they are glad to have been born Negative life assessment: About 10% - He says a minority report their lives have been worse than zero Nonprofit scaling example: $150 million - Future Fund reportedly scaled from zero to about this amount in six months Donation norm: 10% - Giving What We Can encourages members to give at least this share of income Climate policy example: One penny per ton - He suggests an early, gradually rising carbon tax as a gentle, scalable intervention AI alignment outlook: 10 years - He says lack of meaningful progress in a decade would be a major warning sign for the field
Pivotal Quotes: "long-termism is the view that we should be doing much more to protect the interests of future generations than we're currently doing today." — Will MacAskill: His core definition of the philosophy under discussion "think long-term or act now, or like near-term risks with long-term consequences." — Will MacAskill: A concise formulation of his preferred practical framing "if human beings go extinct, we're never coming back from that." — Will MacAskill: Why extinction risk is uniquely important under long-termism
Implications: Listeners are urged to treat future welfare as a major moral priority and to support work on pandemic preparedness, AI safety, and values shaping. The episode also suggests individuals can contribute through donations, careers, and better forecasting, while institutions should invest earlier in preventing catastrophic risks.