EA Forum Podcast (All audio)

This is the first post in a sequence about how I think about and break down my research process. Post 2 is coming soon. Thanks to Oli Clive-Griffin, Paul Bogdan, Shivam Raval and especially to Jemima Jones for feedback, and to my co-author Gemini 2.5 Pro - putting 200K tokens of past blog posts and a long voice memo in the context window is OP.Introduction Research, especially in a young and rapidly evolving field like mechanistic interpretability (mech interp), can often feel messy, confusing, and intimidating. Where do you even start? How do you know if you're making progress? When do you double down, and when do you pivot? These are far from settled questions, but I’ve supervised 20+ papers by now, and have developed my own mental model of the research process that I find helpful. This isn't the definitive way to do research (and I’d love [...] ---Outline:(00:36) Introduction(03:28) The key stages(03:35) Ideation (Stage 0): Choose a problem(04:14) Exploration (Stage 1): Gain surface area(06:55) Understanding (Stage 2): Test Hypotheses(09:17) Distillation (Stage 3): Compress, Refine, Communicate--- First published: April 26th, 2025 Source: https://forum.effectivealtruism.org/posts/hmBPqApDXvhLzbiFt/how-i-think-about-my-research-process-explore-understand --- Narrated by TYPE III AUDIO.

This is a link post. From Rachel Glennerster and Siddhartha Haria of the Center for Global Development. Glennerster is former chief economist of UK's Department for International Development, former head of JPAL, and a 10% pledger with Giving What We Can. "Projects should do one thing on a large scale. The cost per person of implementing a project falls sharply the more people it covers because of the high fixed costs of design, setup, and oversight." "Given the reality of smaller aid budgets, it is imperative we use them to best effect. We need to focus on a select number of simple, cost-effective, large-scale interventions. This will require reforming the review process and making better use of fewer staff, but it will make sure more of our aid gets to those who need it and make it easier for politicians and the public to understand what they are getting [...] --- First published: April 25th, 2025 Source: https://forum.effectivealtruism.org/posts/sgYjJyjoETxGv659i/rachel-glennerster-and-siddhartha-haria-on-simplifying-aid Linkpost URL:https://www.cgdev.org/blog/radical-simplification-practical-way-get-more-out-limited-foreign-assistance-budgets --- Narrated by TYPE III AUDIO.

Another year, another World Malaria Day. WHO reports that an estimated 263 million cases and 597 000 malaria deaths occurred worldwide in 2023, with 95% of the deaths occuring in Africa.What's still the case: Malaria is still one of the top five causes of death for children under 5. The Our World In Data page on Malaria is still a fantastic resource to learn about Malaria.What's different this year:NB- this post is far from comprehensive. I'd appreciate people adding more information about the development of insecticide resistance in mosquitoes, or the speed of the malaria vaccine roll-outs in the comments. Malaria vaccines. We are now over a year into the launch of routine malaria vaccinations in Africa. GAVI has reported that 12 million doses of vaccines have been delivered to 17 countries. There are also very positive signs from the pilot, which ran from 2019-23 [...] ---Outline:(00:27) What's still the case:(00:42) What's different this year:(00:58) Malaria vaccines.(01:48) Cuts to foreign aid--- First published: April 25th, 2025 Source: https://forum.effectivealtruism.org/posts/kbsxHPyw7AZJWxniA/world-malaria-day-2025 --- Narrated by TYPE III AUDIO.

This is a crosspost for The Case for Insect Consciousness by Bob Fischer, which was originally published on Asterisk in January 2025. [Subtitle.] The evidence that insects feel pain is mounting, however we approach the issue. For years, I was on the fence about the possibility of insects feeling pain — sometimes, I defended the hypothesis;[1] more often, I argued against it.[2] Then, in 2021, I started working on the puzzle of how to compare pain intensity across species. If a human and a pig are suffering as much as each one can, are they suffering the same amount? Or is the human's pain worse? When my colleagues and I looked at several species, investigating both the probability of pain and its relative intensity,[3] we found something unexpected: on both scores, insects aren’t that different from many other animals. Around the same time, I started working with an [...] ---Outline:(02:20) Uncomfortable studies(04:12) Mere nociception?(09:14) Don't move the goalposts(10:33) Studying sentience(18:38) Abiding doubt, and some replies(22:42) The path to considerationThe original text contained 9 footnotes which were omitted from this narration. --- First published: April 24th, 2025 Source: https://forum.effectivealtruism.org/posts/2RdYDcwrnvdCn2SbK/the-case-for-insect-consciousness --- Narrated by TYPE III AUDIO. ---Images from the article:Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.

A note on status: The following post is Wild Animal Initiative's first foray into thinking about the intersection of transformative AI and wild animal welfare. I’m not an AI safety expert, but I’ve been largely familiar with the main ideas since 2012. I’ve done my best to read around the issues and asked several more AI-informed folks to give feedback on this draft; however, our ideas at Wild Animal Initiative remain very early days so please take this all with a grain of salt. We welcome additional feedback on how we should be preparing for potential societal changes due to AI. Please feel free to contact strategy@wildanimalinitiative.org to share suggestions or thoughts; I’ll also be engaging with comments on this post.Executive Summary Many in the AI safety community believe transformative AI (TAI; defined here as AI tools with cognitive capabilities surpassing highly trained humans) could arrive within [...] ---Outline:(00:16) Executive Summary(02:15) Transformative AI and wild animals: An exploration.(03:22) When will TAI get here, and what will it bring?(03:46) When will TAI get here?(05:36) What will TAI bring?(07:18) How will a technological & manufacturing explosion affect wild animal welfare science?(07:25) TAI and the strategic importance of academic fields.(12:41) Shortened impact timelines for wild animal welfare science(18:41) Asymmetric acceleration and research priorities(27:27) How does TAI affect WAW directly?(29:21) Space exploration(31:55) Misalignment(34:49) Abundance(38:07) Unknown unknowns(39:00) Summary(39:37) Summary of takeaways for wild animal welfare science(40:43) Summary of takeaways for the wild animal welfare community in general(42:00) Closing thoughts(44:25) Acknowledgments(45:03) Appendix(45:06) TAI and wild animal welfare science: potential effects.(45:38) Grand challenges(45:44) Issues of lower relevance(48:03) Issues of uncertain sign--- First published: April 24th, 2025 Source: https://forum.effectivealtruism.org/posts/zXhxagQKC6kxPM2Kn/transformative-ai-and-wild-animals-an-exploration --- Narrated by TYPE III AUDIO.

[Cross-posted from my Substack here] If you spend time with people trying to change the world, you’ll come to an interesting conundrum: Various advocacy groups reference previous successful social movements as to why their chosen strategy is the most important one. Yet, these groups often follow wildly different strategies from each other to achieve social change. So, which one of them is right? The answer is all of them and none of them. This is because many people use research and historical movements to justify their pre-existing beliefs about how social change happens. Simply, you can find a case study to fit most plausible theories of how social change happens. For example, the groups might say: Repeated nonviolent disruption is the key to social change, citing the Freedom Riders from the civil rights Movement or Act Up! from the gay rights movement. Technological progress is what drives improvements [...] The original text contained 1 footnote which was omitted from this narration. --- First published: April 24th, 2025 Source: https://forum.effectivealtruism.org/posts/kACcdhLDdWb9ZPG9L/why-you-can-justify-almost-anything-using-historical-social --- Narrated by TYPE III AUDIO. ---Images from the article:Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.

This is a link post.About two months ago, @finm and I recorded a discussion about my now-published piece on “AI tools for Existential Security”[1] (for the podcast Fin has been running at Forethought). I’d never done something like this before, and liked the result a lot more than I thought I would![2]I'm sharing a quick transcript here in case anyone appreciates section headers, hyperlinks, etc.. Please note that this is very rough; it's an unpolished, not-double-checked “MVP” edit of an AI-generated transcript, and will definitely have mistakes.[3] Here's the episode itself. You should also be able to find it on most podcast apps. The discussion loosely follows the structure of the piece it focuses on. My thinking on some of what we discuss here has changed a bit, but I'd still say most of this (sometimes with somewhat different emphasis, framings, etc.) and I wanted to quickly get [...] ---Outline:(01:41) Transcript(01:44) Elevator pitch & how this relates to d/acc and def/acc(05:24) Which kinds of AI applications seem most exciting?(08:08) Getting more specific on applications & discussing adoption(10:49) AI negotiation tools, privacy-preserving information-sharing(19:10) Cybersecurity(22:17) Things that might be possible as AI capabilities ramp up(24:25) Win-win policy advice(28:26) Why these clusters of applications?(31:17) What about market effects (and other dynamics that might make it hard to make a counterfactual difference)?(40:57) Methods: how can we accelerate beneficial AI tools?(51:01) Applications that could be doable with today's models (or generally very soon)(54:52) Why not focus on slowing risk-increasing things down?(57:55) A pre-mortem(59:56) (How) many more people should be working on this?(01:04:04) Other upshots(01:04:27) Speeding up automation of your area, using AI tools more(01:06:48) Preparing for a world with much more abundant cognition(01:11:27) Getting ready to help with automation(01:13:04) Applications we're personally excited for(01:15:19) Key open questions?The original text contained 6 footnotes which were omitted from this narration. --- First published: April 21st, 2025 Source: https://forum.effectivealtruism.org/posts/Bj5naijMQFqoHFMRA/podcast-on-ai-tools-for-existential-security-transcript Linkpost URL:https://pnc.st/s/forecast/30c70e39/ai-tools-for-existential-security-with-lizka-vaintrob- --- Narrated by TYPE III AUDIO. ---Images from the article:Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.

This is a link post. Converting to a for-profit model would undermine the company's founding mission to ensure AGI "benefits all of humanity," argues new letter This is the full text of a post from Obsolete, a Substack that I write about the intersection of capitalism, geopolitics, and artificial intelligence. I’m a freelance journalist and the author of a forthcoming book called Obsolete: Power, Profit, and the Race to Build Machine Superintelligence. Consider subscribing to stay up to date with my work. Don’t become a for-profit. That's the blunt message of a recent letter signed by more than 30 people, including former OpenAI employees, prominent civil-society leaders, legal scholars, and Nobel laureates, including AI pioneer Geoffrey Hinton and former World Bank chief economist Joseph Stiglitz. Obsolete obtained the 25-page letter, which was sent last Thursday to the attorneys general (AGs) of California and Delaware, two officials with the power to [...] ---Outline:(00:15) Converting to a for-profit model would undermine the companys founding mission to ensure AGI benefits all of humanity, argues new letter(02:12) Nonprofit origins(05:18) Nobel opposition(06:34) Contradictions(09:20) Justifications(12:24) No sale price can compensate(13:58) An institutional test(15:44) Appendix: Quotes from OpenAI's leaders over the years--- First published: April 23rd, 2025 Source: https://forum.effectivealtruism.org/posts/qHhkSR2EGcQBqgpS2/openai-alums-nobel-laureates-urge-regulators-to-save-company Linkpost URL:https://garrisonlovely.substack.com/p/breaking-openai-alums-nobel-laureates --- Narrated by TYPE III AUDIO. ---Images from the article:Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.

I'm in the process of setting up The Jeremy Coller Centre for Animal Sentience at the LSE, and I have two new job ads out: Research Officer in Animal Sentience (Animals and AI) - this is an open-ended, research-focused postdoctoral role looking at the impacts AI is having on non-human animals, ways of promoting beneficial uses, and ways of ameliorating harmful uses. Centre Manager (Impact and Administration) - this person will work with me to run the Centre on a day-to-day basis and work towards its impact goals. Both jobs are £42,679-£51,000pa, and based at the LSE in London, with the possibility of hybrid working. Anyone can apply from anywhere in the world. We can usually sponsor visas for successful candidates. I'm looking for people with a sincere commitment to the Centre's mission! You can read about that mission over here. Please get in touch if this might [...] --- First published: April 23rd, 2025 Source: https://forum.effectivealtruism.org/posts/MJRZaWvgfWbNw6JTn/two-jobs-at-the-jeremy-coller-centre-for-animal-sentience --- Narrated by TYPE III AUDIO.

Crosspost from my blog that you should totally check out! If I were an animal, I would be a bug. Statistically. —David Sposito. C.S. Lewis once said, “Christianity, if false, is of no importance, and if true, of infinite importance. The only thing it cannot be is moderately important.” In other words, the thing one cannot think is that Christianity is only a bit important—that following the way of Christ should be part of one's life, but not a big part. You should not be the sort of person who goes to church once a month, occasionally feels contrite about their sins, but basically ignores Christianity most of the time. This is how I feel about insect suffering. I wrote an article recently about insects. In short, I argued that there are so many insects that their suffering is the biggest deal in the world. Most insects live [...] --- First published: April 23rd, 2025 Source: https://forum.effectivealtruism.org/posts/RhcP8TkCMbeRHH8ay/untitled-draft-hrbq --- Narrated by TYPE III AUDIO.