Effective altruism
Using evidence and reason to work out how to benefit others as much as possible, and taking action on that basis.
Most of us want to help other people. Far fewer of us stop to ask how much our helping actually accomplishes — and the honest answer is that the difference between options is enormous. Two charities working on the same problem, with the same budget, can differ in impact by a factor of ten or more. Where you point your career can matter even more than that.
Effective altruism is what you get when you take that question seriously: not just are we doing good? but are we doing as much good as we could be? It is a question, not a fixed set of answers, and reasonable people in this community disagree sharply about the answers.
The core ideas
- Impartiality
- Everyone's wellbeing counts, regardless of where they live, what they look like, or when they are born. Distance is not a moral discount rate.
- Scale, neglectedness, tractability
- Problems differ hugely in how many they affect, how much attention they already get, and how much progress an extra person can make. Looking at all three tends to point somewhere unexpected.
- Evidence and reasoning
- Good intentions do not guarantee good outcomes. Where we can measure, we should. Where we cannot, we should reason carefully and stay honest about our uncertainty.
- Willingness to change your mind
- Conclusions are provisional. The point of the exercise is to end up somewhere better, not to defend where you started.
Where the attention tends to go
Applying those ideas has led a lot of people in this community towards a handful of areas. None of these is an official position — they are just where the arguments have taken many of us so far.
- Global health and development. Some of the best-evidenced interventions in the world — treating malaria, vitamin supplementation, direct cash transfers — save or transform lives for remarkably little money.
- Animal welfare. Tens of billions of animals live in factory farms. Very little money goes towards changing that, which makes the marginal contribution unusually large.
- Risks to humanity's long-term future. Pandemics, advanced AI, nuclear conflict: low-probability, extremely high-stakes, and badly under-resourced relative to what is at stake.
- Improving decision-making and institutions. Better forecasting, better policy, better research — the things that make progress on everything else more likely.
The criticisms are worth taking seriously
We would rather point at these than pretend they do not exist. People argue that effective altruism over-weights what is measurable, that it leans too hard on speculative long-term reasoning, that its funding is concentrated in too few hands, and that the collapse of FTX in 2022 exposed real problems with how the movement handled money and status. Some of that criticism has landed and changed things.
You are not signing up to a doctrine by coming to our events. Some of the most useful people in this community are the ones who think large parts of it are wrong and say so.
Want to dig in properly? Our six-week Introductory Program works through these ideas and their strongest objections, in a small group, with someone to argue with. It is free.
Reading elsewhere
- effectivealtruism.org — the general introduction
- 80,000 Hours — research on high-impact careers
- GiveWell — in-depth charity evaluation
- EA Forum — where the arguments actually happen