EA Forum Podcast (All audio)

Comments from the banner on the frontpage will also be posted below, but you should also feel free to post here any time.

This thread is for discussing ways the world is getting better. I considered making this an “EA wins in 2024” banner, but ultimately, I decided that was a bit too restrictive. However, I’d still love to see examples of ways the EA community is making the world better. To anyone that needs to hear it — it's not bragging, it's sharing useful and motivating information.

Last time we ran this banner, there wasn’t a discussion thread. This meant there was no room for caveats or pushback (horror!). This time, I’d love to see discussions springing out of the comments people make on the banner. However, be aware that these comments will be about things that people find motivating/ exciting, so consider that when you [...]

---

First published:

January 20th, 2025

Source:

https://forum.effectivealtruism.org/posts/wrThDRyyT9zvqCYsn/ways-the-world-is-getting-better-discussion-thread

---

Narrated by TYPE III AUDIO.

We’re developing an AI-enabled wargaming-tool, grim, to significantly scale up the number of catastrophic scenarios that concerned organizations can explore and to improve emergency response capabilities of, at least, Sentinel.

Table of Contents

1. How AI Improves on the State of the Art
2. Implementation Details, Limitations, and Improvements
3. Learnings So Far
4. Get Involved!

How AI Improves on the State of the Art

In a wargame, a group dives deep into a specific scenario in the hopes of understanding the dynamics of a complex system and understanding weaknesses in responding to hazards in the system. Reality has a surprising amount of detail, so thinking abstractly about the general shapes of issues is insufficient. However, wargames are quite resource intensive to run precisely because they require detail and coordination.

Eli Lifland shared with us some limitations about the exercises his team has run, like at The Curve conference:

1. It took [...]

---

Outline:

(00:23) How AI Improves on the State of the Art

(02:33) Implementation Details, Limitations, and Improvements

(04:35) Some Learnings So Far

(08:36) Get Involved!

---

First published:

January 18th, 2025

Source:

https://forum.effectivealtruism.org/posts/dwMYQ2SkkNB7YXmKo/scaling-wargaming-for-global-catastrophic-risks-with-ai

---

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.

In 2023[1] GiveWell raised $355 million - $100 million from Open Philanthropy, and $255 million from other donors.

In their post on 10th April 2023, GiveWell forecast the amount they expected to raise in 2023, albeit with wide confidence intervals, and stated that their 10th percentile estimate for total funds raised was $416 million, and 10th percentile estimate for funds raised outside of Open Philanthropy was $260 million.

 10th percentile estimateMedian estimateAmount raisedTotal$416 million$581 million$355 millionExcluding Open Philanthropy$260 million$330 million$255 million

Regarding Open Philanthropy, the April 2023 post states that they "tentatively plans to give $250 million in 2023", however Open Philanthropy gave a grant of $300 million to cover 2023-2025, to be split however GiveWell saw fit, and it used $100 million of that grant in 2023.

However for other donors I'm not sure what caused the missed estimate

Credit to 'Arnold' on GiveWell's December 2024 Open Thread for [...]

The original text contained 2 footnotes which were omitted from this narration.

---

First published:

January 19th, 2025

Source:

https://forum.effectivealtruism.org/posts/RdbDH4T8bxWwZpc9h/givewell-raised-less-than-its-10th-percentile-forecast-in

---

Narrated by TYPE III AUDIO.

It seems to me that we have ended up in a strange equilibrium. With one hand, the Western developed nations are taking actions that have obvious deleterious effects on developing countries... With the other hand, we are trying (or at least purport to be trying) to help developing countries through foreign aid... Probably the strategy that we as the West could be doing, is to not take these actions that are causing harm. That is, we don't need to "fix" anything, but we could stop harming developing countries.

—Nathan Nunn, Rethinking Economic Development

EAs typically think about development policy through the lens of “interventions that we can implement in poor countries.” The economist Nathan Nunn argues for a different approach: advocating for pro-development policy in rich countries. Rather than just asking for more and better foreign aid from rich countries, this long-distance development policy[1] goes beyond normal aid-based development [...]

---

Outline:

(01:17) The argument

(01:21) Do less harm

(05:17) Do some good

(06:54) Isn't this just aid advocacy?

(07:42) Can we do anything about this?

(11:35) Travel visa reforms

(13:50) Consumer purchasing power

(16:16) Payment processing for remittances

(17:53) What would we need to do to make this work?

(20:03) Conclusion

The original text contained 1 footnote which was omitted from this narration.

---

First published:

January 18th, 2025

Source:

https://forum.effectivealtruism.org/posts/j5um9ZhyDCiDXaDqd/long-distance-development-policy

---

Narrated by TYPE III AUDIO.

Ambitious Impact (AIM, formerly Charity Entrepreneurship)[1] is collecting ideas for promising interventions and new charities in animal welfare, global health, and development that have climate change co-benefits (mitigation or adaptation), or vice versa.[2] Specifically, we're looking for ideas that:

• Are cost-effective in the field of global health and development because it improves climate change adaptation.
• Are cost-effective in the field of global health and development, and are also a cost-effective way to reduce greenhouse gas emissions.
• Are cost-effective in the field of animal welfare (farmed or wild), and are also a cost-effective way to reduce greenhouse gas emissions.

The end-goal is to launch new charities that improve the cost-effectiveness of climate funding by achieving more emissions reductions and by increasing co-benefits for humans and/or animals.

How you can help

As part of the first stage of our research process into effective interventions in this cause area, we're welcome [...]

The original text contained 4 footnotes which were omitted from this narration.

---

First published:

January 17th, 2025

Source:

https://forum.effectivealtruism.org/posts/hiMgu748momCycCGo/call-for-suggestions-climate-change-charity-ideas-with

---

Narrated by TYPE III AUDIO.

Hello!

It seems to me that the EA community leans towards progressive or liberal political ideologies. This feels especially pertinent within animal advocacy, where moral and cultural disagreements often create barriers to broader acceptance. However, I think it's an analogous problem within EA and so feel free to weigh in even if animals aren’t your primary concern.

If we agree that engaging more conservatives is desirable, how might we achieve this? Here are a few ideas:

1. Highlight conservative-friendly interventions: Focus on initiatives that align with conservative priorities, such as promoting free-market solutions to factory farming (e.g., supporting cultured meat startups) or emphasizing the health benefits of plant-based diets.

2. Engage conservative leaders: Collaborate with conservative thought leaders, policymakers, and organizations to bridge ideological divides and promote EA principles in ways that resonate with their audiences.

3. Encourage open dialogue: Create spaces within EA [...]

---

First published:

January 17th, 2025

Source:

https://forum.effectivealtruism.org/posts/Gg7WTqeBdTXn2ZnMn/should-we-try-harder-to-appeal-to-conservatives

---

Narrated by TYPE III AUDIO.

If you’re visiting Washington DC to learn more about what's happening in effective altruist policy spaces, we at EA DC want to make sure you get the most out of it! EA DC is one of the largest EA networks and we have a lot of amazing people to draw from for help. We have a lot of activity in each major EA cause area and in a broad range of policy careers, so there are a lot of great opportunities to connect and learn about each space!

If you're not visiting DC soon but would still like to connect or learn more about the group you should email us at [email protected] and explore our resource list!

How to get the most out of DC

Fill out our visitor form

Start by filling out our visitor form. We’ll get back to you soon with any resources and connections [...]

---

Outline:

(00:44) How to get the most out of DC

(00:49) Fill out our visitor form

(01:14) Connect with the EA DC network

(01:44) Get added to the EA DC Slack

(02:05) Get hosted for your stay

(02:15) Find an office to work from during the day

(02:42) Be aware of DC professional culture

(04:32) Explore resources on different policy areas EAs in DC are focused on

(05:16) Get recommendations on things to do in DC

(05:48) Some notes on the EA DC network

---

First published:

January 17th, 2025

Source:

https://forum.effectivealtruism.org/posts/tjDAXeeqhts9ZxjmH/how-ea-dc-can-help-you-get-the-most-out-of-visiting

---

Narrated by TYPE III AUDIO.

As I was driving the other day, I saw a group of protestors in front of the local Methodist church.

"God Hates Abortion! Pray to End the Murder!"

Various signs to this effect were being held up triumphantly by the rather old people who had decided that this was their best use of a Tuesday morning.

This got me curious; how big of an issue really is abortion? I realized I didn't know how many abortions happened per year in the U.S. I stopped by the side of the road to look this up, and was flabbergasted to learn that the number of abortion in the U.S. in 2023 was a little over 1 million.

There were 1 million abortions in 2023.

If you're anything like me, that number is probably a little shocking.

For quick reference, there were about 600,000 malaria deaths in 2023. And that's [...]

---

Outline:

(00:42) There were 1 million abortions in 2023.

(02:14) For each fetus aborted, there were 8,650 broiler chickens that were slaughtered.

(02:51) Conclusion

The original text contained 4 footnotes which were omitted from this narration.

---

First published:

January 15th, 2025

Source:

https://forum.effectivealtruism.org/posts/uryJCRWfG6iyvkQ52/how-big-of-a-problem-is-abortion

---

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.

Project for Awesome (P4A) is a charity video contest running from February 11th to February 19th this year (2025). Participants create short videos supporting a specific charity. Afterwards, the public can vote, and the charities with the most votes receive donations. This presents an excellent (and cost-effective, as explained below) opportunity to raise funds for EA charities and promote EA principles to a wider audience.

In recent years, winning charities have received between $14,000 and $38,000 each. In 2024, over $100,000 was distributed among three different EA charities. Videos don’t need to be professionally made but must be submitted by 11:59 AM EST on Saturday, February 8th.

How It Works

1. Create and Submit Videos
   Participants make 1-4 minute videos supporting charities, upload them to YouTube, and submit them via the P4A website by 11:59 AM EST on Saturday, February 8th (earlier submissions are preferable).
2. Voting Period
   Voting takes [...]
   

---

Outline:

(01:01) How It Works

(02:02) Key Statistics

(03:24) How to Get Involved

(05:04) Opportunities for Charities

---

First published:

January 16th, 2025

Source:

https://forum.effectivealtruism.org/posts/tsoQupFEXjv3btJGu/make-a-short-video-for-your-favorite-ea-charity-project-for

---

Narrated by TYPE III AUDIO.

Scaling inference

With the release of OpenAI's o1 and o3 models, it seems likely that we are now contending with a new scaling paradigm: spending more compute on model inference at run-time reliably improves model performance. As shown below, o1's AIME accuracy increases at a constant rate with the logarithm of test-time compute (OpenAI, 2024).

OpenAI's o3 model continues this trend with record-breaking performance, scoring:

• 2727 on Codeforces, which makes it the 175th best competitive programmer on Earth;
• 25% on FrontierMath, where "each problem demands hours of work from expert mathematicians";
• 88% on GPQA, where 70% represents PhD-level science knowledge;
• 88% on ARC-AGI, where the average Mechanical Turk human worker scores 75% on hard visual reasoning problems.

According to OpenAI, the bulk of model performance improvement in the o-series of models comes from increasing the length of chain-of-thought (and possibly further techniques like "tree-of-thought") and improving the chain-of-thought [...]

---

Outline:

(02:52) AI safety implications

(04:04) AGI timelines

(04:55) Deployment overhang

(06:12) Chain-of-thought oversight

(07:45) AI security

(09:10) Interpretability

(10:07) More RL?

(11:31) Export controls

(11:57) Conclusion

The original text contained 4 footnotes which were omitted from this narration.

---

First published:

January 15th, 2025

Source:

https://forum.effectivealtruism.org/posts/bYXjejHrvq65jFL9s/implications-of-the-inference-scaling-paradigm-for-ai-safety

---

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.