What is effective altruism?
Why this matters
You’re here because you want to do good. Most people do. But wanting to help isn’t enough — we also need to figure out how to help effectively.
That’s harder than it sounds. The world is complex. Our intuitions often mislead us. Good intentions sometimes produce bad outcomes. And the difference between a mediocre intervention and an excellent one can be enormous — not 10% better, but 10x or 100x better.
Effective altruism is the project of figuring out how to do the most good, and then actually doing it. Over the next five weeks, you’ll develop practical skills for this: how to measure impact, how to reason well when faced with uncertainty, how to prioritise between problems, and how to make a plan for your own contribution.
This week, we start with foundations: what EA is, and how to reason transparently about doing good.
What is effective altruism? (20 mins)
Before we dive in, let’s get grounded in what effective altruism actually is.
Wikipedia offers a useful starting point: a philosophical and social movement that advocates using evidence and reasoning to figure out how to benefit others as much as possible, and taking action on that basis.
But there’s more to it than this simple definition. Choose one of the following to get a fuller picture:
Option A: Read this chapter (20 mins) — a philosophically rigorous introduction
Option B: Read or listen to this article (8 mins) — a more accessible overview
Pick whichever suits you, but complete one before continuing.
How we engage with each other (10 mins)
As you can see, effective altruism isn’t just a set of ideas — it’s a community of people trying to figure things out together. Therefore, how we engage with each other really matters.
The guiding principles of effective altruism describe the values that hold this community together:
Commitment to others. We take the wellbeing of others seriously — not just people close to us, but anyone we can help. This includes distant strangers, future generations, and nonhuman animals. We’re willing to take significant action to make the world better, even when it’s inconvenient or uncomfortable.
Scientific mindset. We try to base our actions on evidence and reason. We recognise how hard it is to know how to do the most good, so we avoid overconfidence, seek out critiques of our own views, and take alternative perspectives seriously.
Openness. We’re united by shared principles, not a specific cause. If good arguments or evidence show that our current plans aren’t the best way of helping, we change course. We’re open to unusual ideas if the reasoning is sound.
Integrity. Trust, cooperation, and honest information are essential to doing good. We strive to be honest and trustworthy, and to follow rules of good conduct that allow communities to thrive.
Collaborative spirit. We aim to build a friendly, open environment where different approaches can flourish. We treat people with different worldviews, backgrounds, and identities kindly and respectfully.
When EA has fallen short
In the spirit of integrity, it’s important to acknowledge that the EA community hasn’t always lived up to these principles.
FTX and failures of integrity. In 2022, FTX — a cryptocurrency exchange whose founder was closely associated with effective altruism — collapsed amid revelations of fraud. Billions of dollars were lost, including funds pledged to EA organisations. The founder, who had publicly advocated for earning to give and longtermism, was later convicted of fraud.
This wasn’t a failure of EA ideas — it was a failure of integrity by individuals. But it prompted hard questions: Did the community defer too much to major donors? Did the emphasis on impact create blind spots about how resources were acquired? Were there warning signs that people ignored? The EA community has since reflected extensively on these questions. Some conclusions: integrity matters as much as impact. How you do things matters, not just what you achieve. And communities need to maintain accountability even — especially — for high-status members.
Measurability bias. Early EA focused heavily on interventions where impact could be quantified — global health charities with measurable outcomes like “cost per life saved.” This had genuine advantages: it enabled rigorous comparison and helped identify outstanding giving opportunities. But it also meant neglecting important problems that are harder to measure: systemic change, policy reform, and speculative but potentially high-impact work on emerging risks.
Recognising this limitation, EA funders created organisations like Open Philanthropy (since renamed Coefficient Giving) to support high-impact work that doesn’t fit neatly into cost-effectiveness spreadsheets — including policy research, institution-building, and work on catastrophic risks. The scientific mindset means updating when you notice your methods have blind spots.
Lack of diversity. The EA community has remained demographically narrow. The 2024 EA Survey found that 69% of respondents identify as men and 75% as white. People from low- and middle-income countries — where many EA-funded programmes operate — are underrepresented. This matters for several reasons: diverse perspectives improve decision-making, narrow demographics can create blind spots, and a movement that claims to care about everyone should be welcoming to everyone.
Progress has been slow. The community has acknowledged this as a problem and various organisations have taken steps to address it, but demographic composition hasn’t shifted substantially. This remains an area where stated values and actual outcomes are misaligned.
Learning from mistakes publicly. One way the community tries to embody the scientific mindset is by documenting errors openly. Organisations like the Centre for Effective Altruism, 80,000 Hours and GiveWell maintain public “mistakes” pages cataloguing their errors and what they’ve learned. For an independent view, you might also want to check the ‘Other criticisms and controversies’ section of the effective altruism Wikipedia page and click through the references.
What this means for our discussions:
Throughout this course, you’ll share your reasoning with others and receive feedback. A few norms will help this go well:
Be honest about uncertainty. “I’m not sure” and “I don’t know” are valuable contributions. Pretending to be more confident than you are makes it harder for others to help you think.
Seek to understand before critiquing. When someone shares their reasoning, make sure you understand their position before pushing back. Paraphrase what you heard. Ask clarifying questions.
Critique ideas, not people. You can disagree strongly with someone’s reasoning while still treating them with respect. Focus on “here’s why I think the argument doesn’t work” rather than “you’re wrong.”
Be willing to change your mind. The goal isn’t to win arguments — it’s to get closer to the truth. If someone makes a good point, say so. Updating your views is a sign of intellectual honesty, not weakness.
Assume good faith. Everyone in this course wants to do good. When someone says something you find strange or wrong, start by assuming they have reasons you haven’t understood yet.
These norms aren’t just about being nice — though they are that too. They create the conditions for genuine learning. When people feel safe to share half-formed thoughts and admit uncertainty, the whole group thinks better.