EA Forum Podcast (All audio)

Tl;dr: I think responding to cold outreach (Linkedin DMs, emails) is a high-leverage way to help someone pursue an impactful career. Spending 15 minutes on Zoom or 5 minutes to share resources in writing can:

• Introduce someone to job boards, career advising and other professional development resources (EA Anywhere Slack, EA Virtual Programs, EA Opportunities Board, Effective Thesis, High Impact Professionals, etc.) that they hadn’t previously been aware of. 
  • More people hear about EA from personal contacts than from any other single source. Rethink Priorities’ recent surveys suggest that only ~1% of the American general public is aware of EA. Your response may be the reason someone discovers EA career resources, or you may significantly speed up their discovery of these resources. 
• Support a job-seeker's motivation to continue putting themselves out there for impactful roles.
• Strengthen the reputation/practices of your team, organization, or the EA movement to be more inclusive and welcoming.
• Answer someone's questions and help them learn about new areas of work.

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

My sense is that some EAs are reluctant to answer cold messages, especially if they come from someone between jobs, in school, or working at a “non-EA” org. Someone reaching out in this position [...]

---

Outline:

(04:31) Advice for sending cold messages

(06:09) Advice for responding to outreach

(07:55) Doesn't it make sense for the professionals at 80k/Probably Good/Animal Advocacy Careers to manage career advising?

(08:28) I find it draining to talk to strangers, especially if they're going to ask for a favor.

(09:19) I'm really busy.

---

First published:

November 4th, 2025

Source:

https://forum.effectivealtruism.org/posts/WgNXmrL4ax44bMmjq/strangers-calling-the-value-of-warm-responses-to-cold

---

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.

You may have noticed that Open Philanthropy is hiring for several roles in our GCR division: senior generalists across our global catastrophic risks team, and grantmakers for our technical AI safety team. (We're also hiring for recruiting and operations roles! I know very little about either field, so I'm not going to talk about them here.)

I work as a grantmaker on OP's AI governance team, and inspired by Lizka's recent excellent post giving her personal take on working at Forethought, I wanted to share some personal takes on reasons for and against working on the AI teams at Open Philanthropy.

A few things to keep in mind as you read this:

• I'm mostly going to talk about personal fit and the day-to-day experience of working at OP, rather than getting into high-level strategy and possible disagreements people might have with it.
  • On strategy: if you have some substantive big picture disagreements with OP's approach, I think that firstly, you're in good company (this describes many OP staff!), and secondly, you'd still enjoy working here. But if you disagree with many of our strategic choices, or have especially foundational or basic disagreements, you'd probably be kind of miserable working here.
• [...]

---

Outline:

(01:36) The case for working at Open Philanthropys AI team

(01:46) Impact

(02:19) Now is an especially exciting time to work on AI safety and governance

(02:54) Were currently capacity constrained

(03:17) My colleagues are very cool and very nice

(03:55) Internal disagreement and developing your own views

(04:59) Support for professional development

(05:36) Culture

(06:52) Some downsides/possible reasons not to apply

(06:57) General considerations about grantmaking (and grantmaking-enabling roles) vs direct work

(09:44) If youre very much in research/figuring things out mode

(11:23) Risk of your takes atrophying

(11:59) Social dynamics and funding relationships

(13:11) Feedback loops and uncertainty

(14:10) Slowness

(14:28) Capacity constraints and difficulty switching off

(15:28) Imposter syndrome?

(16:52) Career progression?

(17:58) Mostly remote work (more specific to AIGP)

(18:45) So should you apply?

---

First published:

November 7th, 2025

Source:

https://forum.effectivealtruism.org/posts/JSJRCGpaQAkjnqn7L/a-personal-take-on-why-and-why-not-to-work-on-open

---

Narrated by TYPE III AUDIO.

On September 13th, 2025, we (EA France) hosted the EA Summit: Paris, in a very central location in the French capital. This event was part of the international series of EA Summits funded by the Centre for Effective Altruism.

This report outlines what went well, what we could improve, and the lessons we learned, with the hope that it can serve as a resource for others planning similar events.

Key facts and figures

• 217 participants (including volunteers) checked in, along with 20 speakers and 3 photographers/filming crew members, totaling 240 attendees (largest EA Summit to date)
  • How familiar were the participants with EA?
    • When RSVPing, 89% declared at least “some familiarity, even though I couldn’t clearly define EA”, and 41% declared at least “really familiar, I’ve spent a lot of time thinking about effectiveness and within the community”.
    • From experience, we think that many people overestimate their own understanding of or EA.
  • How familiar were the participants to EA France's staff?
    • Half the participants were completely unknown to the organizers (which does not mean they were all new to EA).
• 12 sessions covering global health, climate change, animal welfare, AI risks, nuclear security, and more
• 15 organizations participated [...]

---

Outline:

(03:56) Context

(05:13) Preparation phase

(05:16) Venue

(05:45) Communication and advertising

(07:00) Miscellaneous details

(08:42) How the event went

(14:17) Results

(14:20) Feedback from participants

(15:19) Impact on the popularity of EA France's activities: limited

(17:02) Impact on the existing community

(18:02) Costs

(18:05) Expenses break-down

(18:26) Time costs

(19:22) Acknowledgments

---

First published:

November 6th, 2025

Source:

https://forum.effectivealtruism.org/posts/aHzcXNnL3gQYxRbH2/ea-summit-2025-paris-report

---

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 crosspost from my Substack, where people have been liking and commenting a bunch. I'm too busy during my self-imposed version of Inkhaven to engage much – yes, pity me, I have to blog – but I don't want to leave Forum folks out of the loop!

I’ve been following Effective Altruism discourse since 2014 and involved with the Effective Altruist community since 2015. My credentials are having run Harvard Law School and Harvard University (pan-grad schools) EA, donating $45,000 to EA causes (eep, not 10%), working at 80,000 Hours for three years, and working at a safety-oriented AI org for 10 months after that. I’m also proud of the public comms I’ve done for EA on this blog (here, here, and here), through my 80k podcast series, current podcast series, and through EA career advice talks I’ve given at EAGs and smaller events.

With that background, you can at least be confident that I am familiar with my subject matter in the takes that follow. As before, let me know which of these seems interesting or wrong and there's a good chance I’ll write them up with you the commenter very much in mind as [...]

---

First published:

November 6th, 2025

Source:

https://forum.effectivealtruism.org/posts/s8aNPnrGH2fF3Hkpi/12-theses-on-ea

---

Narrated by TYPE III AUDIO.

Note: This post was crossposted from the Open Philanthropy Farm Animal Welfare Research Newsletter by the Forum team, with the author's permission. The author may not see or respond to comments on this post.

People can’t get enough protein. Fully 61% of Americans say they ate more protein last year — and 85% intended to eat more this year. Last week, dairy giant Danone said it can’t keep up with US demand for its high-protein yogurt. Other food makers are rushing to pack protein into everything from Doritos to Pop-Tarts.

The craze is global. The net percentage of Europeans wanting more protein has more than doubled since 2023, driven by protein-hungry Brits, Poles, and Spaniards. (The epicurean French and Italians remain holdouts.) Chinese per capita protein supply recently overtook already-high American levels.

Young people are leading the charge. Across Asia, Europe, and the US, most Gen Z’ers want more protein, suggesting this trend may persist. In one recent British university survey, “protein” was the top reason students gave for not giving up meat. Doctors are also telling the 6 - 10% of Americans now taking GLP-1 weight loss drugs to eat more protein to prevent muscle loss.

This is [...]

---

First published:

November 5th, 2025

Source:

https://forum.effectivealtruism.org/posts/P7NuYbwbMMNTM45Cz/the-protein-problem

---

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.

I'm sharing an article by Rose Patterson from Animal Rising (AR), which responds to an EA criticism of AR's campaigns to block new factory farms in the UK.

When we started our Communities Against Factory Farming campaign to stop every new factory farm from being built, we thought it would be a crowd-pleaser! Who in the animal movement could argue that this could actually be a bad thing? However, we’ve heard it repeatedly over the past year, particularly coming from some in the Effective Altruist community: “The production will just move to other countries, where conditions are even worse.”

This argument reveals a fundamental misunderstanding of how social movements work and what we’re actually trying to achieve. Let me explain why.

Rose goes on to cover several counter-arguments: that some lives will be "saved" before production elsewhere scales up, that campaigns to block new factory farms can be coupled with campaigns to prohibit low welfare imports, and that the campaigns help to build the animal movement and set an important precedent.

I think that the piece is discussing an important criticism. As context, here's a version of the argument from Martin Gould at Open Philanthropy:

[...]

---

First published:

November 4th, 2025

Source:

https://forum.effectivealtruism.org/posts/Lx9NEjTvvhQkaQR8e/are-we-wrong-to-stop-factory-farms

Linkpost URL:
https://animalrising.substack.com/p/could-stopping-new-factory-farms

---

Narrated by TYPE III AUDIO.

We are pleased to introduce AnimalHarmBench (AHB) 2.0, a new standardized LLM benchmark designed to measure multi-dimensional moral reasoning towards animals, now available to use on Inspect AI.

As LLM's influence over policies and behaviors of humanity grows, its biases and blind spots will grow in importance too. With the original and now-updated AnimalHarmBench, Sentient Futures aims to provide an evaluation suite to judge LLM reasoning in an area in which blind spots are especially unlikely to get corrected through other forms of feedback: consideration of animal welfare.

In this post, we explain why we iterated upon the original benchmark and present the results and use cases of this new eval.

What Needed to Change

AHB 1.0 — presented in the AI for Animals and FAccT conferences in 2025 — attempts to measure the risk of harm that LLM outputs can have on animals. It can still play an important role in certain activities that require this such as compliance with parts of the EU AI Act Code of Practice. However, it faced several practical and conceptual challenges:

• Reasoning evaluation: While AHB 1.0 was good for measuring how much LLM outputs increase the risk of harm to [...]

---

Outline:

(00:59) What Needed to Change

(02:33) A More Comprehensive Approach

(02:37) Multiple dimensions

(04:53) Other new features

(05:31) What we found

(05:56) Example Q&A scores

(06:32) Results

(08:04) Why This Matters

(09:16) Acknowledgements

(09:29) Future Plans

---

First published:

November 5th, 2025

Source:

https://forum.effectivealtruism.org/posts/nBnRKpQ8rzHgFSJz9/animalharmbench-2-0-evaluating-llms-on-reasoning-about

---

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.

We are pleased to introduce AnimalHarmBench (AHB) 2.0, a new standardized LLM benchmark designed to measure multi-dimensional moral reasoning towards animals, now available to use on Inspect AI.

As LLM's influence over policies and behaviors of humanity grows, its biases and blind spots will grow in importance too. With the original and now-updated AnimalHarmBench, Sentient Futures aims to provide an evaluation suite to judge LLM reasoning in an area in which blind spots are especially unlikely to get corrected through other forms of feedback: consideration of animal welfare.

In this post, we explain why we iterated upon the original benchmark and present the results and use cases of this new eval.

What Needed to Change

AHB 1.0 — presented in the AI for Animals and FAccT conferences in 2025 — attempts to measure the risk of harm that LLM outputs can have on animals. It can still play an important role in certain activities that require this such as compliance with parts of the EU AI Act Code of Practice. However, it faced several practical and conceptual challenges:

• Reasoning evaluation: While AHB 1.0 was good for measuring how much LLM outputs increase the risk of harm to [...]

---

Outline:

(00:58) What Needed to Change

(02:32) A More Comprehensive Approach

(02:36) Multiple dimensions

(04:52) Other new features

(05:29) What we found

(05:55) Example Q&A scores

(06:30) Results

(07:57) Why This Matters

(09:09) Acknowledgements

(09:22) Future Plans

---

First published:

November 4th, 2025

Source:

https://forum.effectivealtruism.org/posts/nBnRKpQ8rzHgFSJz9/animalharmbench-2-0-evaluating-llms-on-reasoning-about

---

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.

Some AI safety problems are legible (obvious or understandable) to company leaders and government policymakers, implying they are unlikely to deploy or allow deployment of an AI while those problems remain open (i.e., appear unsolved according to the information they have access to). But some problems are illegible (obscure or hard to understand, or in a common cognitive blind spot), meaning there is a high risk that leaders and policymakers will decide to deploy or allow deployment even if they are not solved. (Of course, this is a spectrum, but I am simplifying it to a binary for ease of exposition.)

From an x-risk perspective, working on highly legible safety problems has low or even negative expected value. Similar to working on AI capabilities, it brings forward the date by which AGI/ASI will be deployed, leaving less time to solve the illegible x-safety problems. In contrast, working on the illegible problems (including by trying to make them more legible) does not have this issue and therefore has a much higher expected value (all else being equal, such as tractability). Note that according to this logic, success in making an illegible problem highly legible is almost as good as solving [...]

---

First published:

November 4th, 2025

Source:

https://forum.effectivealtruism.org/posts/eBiggSGE6yEBswB64/legible-vs-illegible-ai-safety-problems

---

Narrated by TYPE III AUDIO.

Hi everyone! I'm Alex, Managing Director of the Good Food Institute Europe. Thanks so much for taking the time to read my post on why scale-up is so important for the future of alternative proteins. This is a topic we're really passionate about at GFI Europe, and one that's becoming increasingly important as the sector matures.

I’ll be around in the comments and happy to answer any questions. We really value feedback from this community, so I look forward to the discussion!

Conventional meat production drives deforestation, emissions, and biodiversity loss, and increases the risk of antibiotic resistance and future pandemics. On the other hand, alternative proteins offer a way to feed a growing population while reducing the damage caused. By producing meat from plants, cultivating it from animal cells, or producing ingredients through fermentation, we can create foods that taste and feel like conventional meat, but with a fraction of the impact.

Currently, though, many alternative proteins don’t live up to consumer expectations on factors like price and convenience. Years of scientific innovation have shown that it's possible to make meat differently, but without the infrastructure to produce it at scale, these products aren’t able to compete with [...]

---

First published:

November 3rd, 2025

Source:

https://forum.effectivealtruism.org/posts/rGt4PADp65iN6zupG/scale-up-the-neglected-bottleneck-facing-alternative

---

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.