Episode Summary
Executive Summary: Sean Carroll interviews Will MacAskill about long-termism, effective altruism, and how moral reasoning should account for future generations. MacAskill argues that the vast scale of the future, existential risks like pandemics, bioweapons, and AI, and the possibility of value lock-in make near-term choices morally pivotal. They also debate utilitarianism, population ethics, uncertainty, democracy, and practical actions like donating, career choice, voting, and having children.
Main Topics: Long-termism and the moral importance of the future (Priority: 5/5): MacAskill defines long-termism as taking seriously the immense scale of humanity’s future and the possibility that present actions could profoundly shape it. Existential risk and catastrophic threats (Priority: 5/5): The conversation emphasizes pandemics, engineered pathogens, nuclear conflict, and AI as real risks that could truncate civilization or lock in bad outcomes. Consequentialism, utilitarianism, and moral realism (Priority: 4/5): MacAskill explains his consequentialist leanings, his partiality to utilitarian reasoning, and his meta-ethical openness to moral realism versus nihilism. Population ethics and the repugnant conclusion (Priority: 5/5): A major section examines how to compare worlds with different population sizes and well-being levels, including the repugnant conclusion and critical-level views. Uncertainty, forecasting, and policy under risk (Priority: 4/5): They discuss why uncertainty is not an excuse for inaction and how forecasting tools and probability estimates can guide decisions about rare, high-stakes events. Practical action: donations, careers, voting, and family (Priority: 4/5): MacAskill highlights effective altruism strategies—giving, career choice, voting, and potentially having children—as ways to improve both present and future welfare. Democracy and institutional representation of future generations (Priority: 3/5): The episode considers whether democratic institutions underweight future people and explores proposals like ombudspersons and citizens’ assemblies.
Key Arguments: The future matters enormously because there may be thousands, millions, or billions of years of human or post-human civilization ahead, so even small improvements in trajectory can have vast moral significance. Extinction risk is non-negligible and should influence current policy, especially because pandemics, bioweapons, and AI-related failures are increasingly plausible and potentially catastrophic. Discounting the future at a fixed positive rate is morally absurd over long time horizons; a small benefit sooner should not outweigh vastly larger harms later simply because of time. Future people should count morally much like distant present people: uncertainty reduces confidence but does not justify ignoring them. Not all future welfare will be human; what matters is conscious beings and valuable future cultures, whether biological, artificial, or hybrid. Population ethics creates genuine paradoxes; any account must reject or revise some intuitive principle, and MacAskill favors a compromise view such as critical-level approaches rather than simplistic aggregation. Democratic systems structurally underrepresent future generations, so institutions may need explicit mechanisms to defend long-term interests. A practical long-termist program is not mutually exclusive with helping the present; it prioritizes high-leverage causes and also recognizes immediate moral obligations. Even if moral realism is uncertain, MacAskill argues it is rational to act as though morality is real when the alternative is nihilism or indecision.
Data Points: Age of the universe: 13.8 billion years - Used to frame how early humans are in cosmic history. Age of Earth: 4.5 billion years - Part of the timeline showing humanity’s late arrival. First replicators: 3.8 billion years ago - Evolutionary timeline mentioned in the long-view framing. First eukaryotes: 2.7 billion years ago - Continuation of the biological history overview. First animals: About 800 million years ago - Timeline of life leading to humans. Human beings: 300,000 years ago - Placed in context as a very recent species. Agricultural revolution: 12,000 years ago - Shows how recent civilization is. Scientific revolution: A few hundred years ago - Illustrates the extreme recency of rapid progress. Pure time preference (economics): 2% per year - Used as an example of discounting the future in a way MacAskill criticizes. Example of future moral choice: One death in 10,000 years vs. genocide of a million in 11,000 years - Illustrates why simple time discounting is morally problematic. Engineered pathogen extinction risk: About 0.5% total risk of extinction - MacAskill’s rough estimate after considering engineered pathogens and the share leading to extinction. Chance of pandemic killing at least 10 million people (Metaculus, 2016-2026): One in three - Forecasting example used to show how serious pandemic risks can be. Chance of engineered pathogen killing at least 95% of the world’s population: About 0.9% - Metaculus estimate cited in the discussion of biological catastrophe. Overall existential risk in our lifetime: About one in six - MacAskill cites Toby Ord’s estimate and says he does not significantly disagree. Chance of a pandemic much worse than COVID-19 in our lifetimes: One in three, perhaps - Used to argue that the risk is large enough to warrant action. Chance of World War III in our lifetime: At least 25% - Presented as another major source of existential risk. Projected economic growth continuation: 2% per year for 10,000 years - Used to show the absurdity of extrapolating current growth indefinitely. Projected economy size increase: 10^89 times as big - Result of 2% growth over 10,000 years. Atoms within 10,000 light years: 10^67 atoms - Used to highlight physical constraints on indefinite growth. Relative representation of future generations in democracy: 1.1% representation - If future generations outnumber the present by 1000 to 1, current voters are tiny share of affected people. Recommended charitable giving: 10% of income - Giving What We Can’s suggested pledge for effective altruism. Human work life: 80,000 hours - Used to motivate career choice as a major lever for impact.
Pivotal Quotes: "Long-termism really means long-term. Not just thinking, oh, yeah, years, decades into the future, but really for how long we might live." — Will MacAskill: Defines the core philosophy of long-termism. "The future is going to be wild, and we should really at least be modally thinking: what are the things we can do that might be impacting the long term, and how can we make them go better." — Will MacAskill: Summarizes the practical imperative of long-term thinking. "I call this the horror of effective altruism, the absolute horror of the world we are in at the moment." — Will MacAskill: Describes the painful trade-offs involved in choosing among high-impact causes and donations.
Implications: Listeners are urged to treat future risk reduction as a serious moral priority, not sci-fi speculation. The episode suggests that careers, donations, governance, and technology policy should be evaluated by their long-run effects on civilization.
About Sean Carroll MindScape
Ever wanted to know how music affects your brain, what quantum mechanics really is, or how black holes work? Do you wonder why you get emotional each time you see a certain movie, or how on earth video games are designed? Then you’ve come to the right place. Each week, Sean Carroll will host conversations with some of the most interesting thinkers in the world. From neuroscientists and engineers to authors and television producers, Sean and his guests talk about the biggest ideas in science, ...