In one of the best summaries of effective altruism to date, Ajeya Cotra draws from her personal experience and lays out the movement’s key concepts and core beliefs. She discusses emotional reactions to the unfairness of the world, the need to prioritize the interventions that can help the most people (or animals), a framework for determining which interventions are most promising, and goes into depth on a few key causes in particular. If you only have time to watch one video about EA, this is a great choice.
“The scientific and industrial revolutions transformed both our understanding of the world, and our ability to alter it. What we need is an ethical revolution, so we can work out: how do we use this tremendous bounty of resources to improve the world?”
If you see a child in mortal peril on the street, surely you have an ethical obligation to help. Why, then, would that obligation not extend to the thousands of children dying every day from preventable diseases? Peter Singer argues that it does, and explores how best to help.
Oxford University's Owen Cotton-Barratt discusses how effective altruists can improve the world. He explores a series of key effective altruist concepts at an intermediate level, such as heavy-tailed distributions, diminishing marginal returns, and comparative advantage.
In this 6 minute talk, Beth Barnes notes that many things about our world aren’t built to maximize what we truly value most. But she suggests - and expands on - a seemingly simple solution: “Give more, give more effectively, and encourage others to do the same."
It can be tempting to adopt a “soldier mindset” where winning arguments is what matters, and opposing evidence must be neutralized with extreme prejudice. But if we want to understand the world well, it pays to have a “scout mindset” and observe the terrain with an open mind.
Because of the potential size of the future, the most important thing about our actions today might be their impact on future generations. This simple-sounding idea has some surprising moral implications.
Mesa-optimization is a recent idea in AI safety, and an important one: what happens when a system devises its own subsystem to pursue a slightly different - but seemingly correlated - objective to the one its designer intended? Robert Miles goes in depth.
Edouard Mathieu presents on the lessons of data communication in the COVID-19 pandemic, followed by a fireside chat with Hannah Ritchie and a Q&A on Our World in Data discussing how it fits with the effective altruism framework.