EA Forum Podcast (All audio)

Imagine you are a junior advisor to the boss of some major EA longtermism-sympathetic org (OpenPhil, CEA, 80K, etc). You are tasked with reading the Essays on Longtermism compilation, and collating any novel insights that could significantly change what we should be doing. That is, we want essays that make ‘big, if true’ claims, and present interesting arguments for them. To be clear, this is a high bar! Novelty is hard, and action-relevance is even harder for orgs that have already thought a lot about these issues.

I claim that there is a surprising dearth of such insights in the Essays (Section 1 and Appendix). I then consider why this might be the case (Section 2), and what this could mean for global priorities research generally (Section 3).

1. Where are all the new, important insights?

Disclaimers first:

• I found several of the essays interesting, and by more [...]

---

Outline:

(00:58) 1. Where are all the new, important insights?

(04:43) 2. Hypotheses

(06:18) 3. Recommendations

(08:58) Appendix: critical reviews of the essays

---

First published:

October 18th, 2025

Source:

https://forum.effectivealtruism.org/posts/2uxdQKhGsLCJFwHck/are-longtermist-ideas-getting-harder-to-find

---

Narrated by TYPE III AUDIO.

This is a Draft Amnesty Week draft. It may not be polished, up to my usual standards, fully thought through, or fully fact-checked. 

Commenting and feedback guidelines:
I'm posting this to get it out there and hopefully elicit discussion but am pretty uncertain about this post as I am not that familiar with global development and am also relatively new to AI safety. So I could be overlooking some important factors.

I would welcome any feedback, particularly from those with experience working with developing countries.

Summary

• TAI's impact on developing countries[1] is likely to look quite different from its impact on developed countries. Labour displacement will be slower, and there will be less fiscal pressures on developing governments. But impacts will be uneven, and countries that rely heavily on service exports and remittances are likely to lose out in the near term.

• Global inequality is likely to [...]

---

Outline:

(00:13) Summary

(01:26) Assumptions and caveats

(02:28) TAI's impact on developing countries is likely to look quite different

(02:34) Less labour displacement

(04:28) Less government fiscal pressures

(05:43) But impacts will be uneven

(07:21) Global inequality is likely to increase, but this won't necessarily worsen global welfare

(09:20) TAI could produce global public goods

(11:32) Weakening influence of liberal democracies

---

First published:

October 18th, 2025

Source:

https://forum.effectivealtruism.org/posts/4apMQAnv5wuystcfx/how-i-expect-tai-to-impact-developing-countries

---

Narrated by TYPE III AUDIO.

DMT, the smallest microdose of maybe 5mg with a vape pen stops the worst pain known, in literally 10–20 seconds. Acid and mushrooms as well, but they take time to come up. Even the smallest sub-perceptual dose of DMT stops the pain.

- Yiftach Yerushalmy, cluster headache patient

One inhalation [of DMT] will end the attack for most people. Everybody is reporting the exact same thing. […] It could end that attack in less than a minute. […] You can take one inhalation, you can wait 30 seconds, and if that cluster is not gone completely, then you know it's time to take another inhalation. You don't have to wait 2h into a psilocybin trip.

- Bob Wold, president of Clusterbusters and cluster headache patient

I use the word game changer because a lot of times our attacks will come at night after we go to sleep. [...]

---

Outline:

(03:14) How DMT is used for cluster headaches

(07:57) The evidence

(12:10) How and why would DMT work to treat cluster headaches?

(14:06) Challenges

(16:03) What is needed next

(18:24) Appendix: Bob Wold's Pocket Guide to Cluster Headaches: DMT

(21:18) DMT Vape pens and cartridges. For use in treating cluster headaches.

(21:25) Cartridges

(25:13) Inhalations

(31:06) DMT - Alternative Source - Delivery System

(32:07) General notes on psychedelic treatments

(33:25) When to begin dosing with psychedelics

---

First published:

October 18th, 2025

Source:

https://forum.effectivealtruism.org/posts/x8P8EGnujSZm6fyMH/emerging-evidence-on-treating-cluster-headaches-with-dmt

---

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.

Note: This is a somewhat rough post published as part of Draft Amnesty Week 2025.

I very, very strongly believe there's essentially no chance of artificial general intelligence (AGI) being developed within the next 7 years.[1] Previously, I wrote a succinct list of reasons to doubt near-term AGI here. For example, it might come as a surprise that around three quarters of AI experts don’t think scaling up large language models (LLMs) will lead to AGI!

For a more technical and theoretical view (from someone with better credentials), I highly recommend this recent video by an AI researcher that makes a strong case for skepticism of near-term AGI and of LLMs as a path to AGI:

In this post, I will give more detailed arguments about why I think near-term AGI is so unlikely and why I think LLMs won't scale to AGI.

Clarification of terms

By essentially no chance, I mean less than the chance of Jill Stein running as the Green Party candidate in 2028 and winning the U.S. Presidency. Or, if you like, less than the chance Jill Stein had of winning the presidency in 2024. I mean it's an incredible long shot [...]

---

Outline:

(01:10) Clarification of terms

(02:37) Some recent AI predictions that turned out to be wrong

(03:40) Benchmarks do not test general intelligence

(05:38) AI in the economy

(07:38) AI vs. puzzles

(08:15) Problems with scaling up LLMs

(09:35) Agentic AI is unsuccessful

(10:50) Conclusion

---

First published:

October 16th, 2025

Source:

https://forum.effectivealtruism.org/posts/sQSCqpm9Ymwiu8rdb/agi-by-2032-is-extremely-unlikely

---

Narrated by TYPE III AUDIO.

Note: This is a somewhat rough post published as part of Draft Amnesty Week 2025.

I very, very strongly believe there's essentially no chance of artificial general intelligence (AGI) being developed within the next 7 years.[1] Previously, I wrote a succinct list of reasons to doubt near-term AGI here. For example, it might come as a surprise that around three quarters of AI experts don’t think scaling up large language models (LLMs) will lead to AGI!

For a more technical and theoretical view (from someone with better credentials), I highly recommend this recent video by an AI researcher that makes a strong case for skepticism of near-term AGI and of LLMs as a path to AGI:

In this post, I will give more detailed arguments about why I think near-term AGI is so unlikely and why I think LLMs won't scale to AGI.

Clarification of terms

By essentially no chance, I mean less than the chance of Jill Stein running as the Green Party candidate in 2028 and winning the U.S. Presidency. Or, if you like, less than the chance Jill Stein had of winning the presidency in 2024. I mean it's an incredible long shot [...]

---

Outline:

(01:07) Clarification of terms

(02:34) Some recent AI predictions that turned out to be wrong

(03:38) Benchmarks do not test general intelligence

(05:35) AI in the economy

(07:36) AI vs. puzzles

(08:12) Problems with scaling up LLMs

(09:32) Agentic AI is unsuccessful

(10:48) Conclusion

---

First published:

October 16th, 2025

Source:

https://forum.effectivealtruism.org/posts/sQSCqpm9Ymwiu8rdb/agi-by-2032-is-extremely-unlikely

---

Narrated by TYPE III AUDIO.

Note: This is a fairly rough post I adapted from some comments I recently wrote that I worked hard enough on that I figured I should probably make them into a post. So, although this post is technically not a draft, it isn't written how I would write a post — it's less polished and more off the cuff. If you think I should remove the Draft Amnesty tag, please say so, and I will!

I very, very strongly believe there's essentially no chance of artificial general intelligence (AGI) being developed within the next 7 years.[1] Previously, I wrote a succinct list of reasons to doubt near-term AGI here. For example, it might come as a surprise that around three quarters of AI experts don’t think scaling up large language models (LLMs) will lead to AGI!

For a more technical and theoretical view (from someone with better [...]

---

Outline:

(01:23) Clarification of terms

(02:50) Some recent AI predictions that turned out to be wrong

(03:54) Benchmarks do not test general intelligence

(05:51) AI in the economy

(07:51) AI vs. puzzles

(08:28) Problems with scaling up LLMs

(09:45) Agentic AI is unsuccessful

(11:01) Conclusion

---

First published:

October 16th, 2025

Source:

https://forum.effectivealtruism.org/posts/sQSCqpm9Ymwiu8rdb/agi-by-2032-is-extremely-unlikely

---

Narrated by TYPE III AUDIO.

This is a Draft Amnesty Week draft. It may not be polished, up to my usual standards, fully thought through, or fully fact-checked. 

Commenting and feedback guidelines:
Keep one and delete the rest (or write your own):

1. I'm posting this to get it out there. I'd love to see comments that take the ideas forward, but criticism of my argument won't be as useful at this time.
2. This draft lacks the polish of a full post, but the content is almost there. The kind of constructive feedback you would normally put on a Forum post is very welcome.
3. This is a Forum post that I wouldn't have posted without the nudge of Draft Amnesty Week. Fire away! (But be nice, as usual)

Introduction

So, today I heard about the Consumer Reports study to do with lead in Huel (and other drinks). And I was pretty concerned about this as I [...]

---

Outline:

(00:12) Introduction

(00:45) Safety thresholds of heavy metals

(01:33) Huel tests and stats vs CR

(01:37) Consumer Reports

(03:35) Huel

(04:33) Chocolate

(05:11) Lead levels in Fruit and Veg

(07:07) Conclusion

(07:28) Further conclusions:

---

First published:

October 15th, 2025

Source:

https://forum.effectivealtruism.org/posts/HLbATPd5GcQSEkndG/heavy-metals-in-huel

---

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 Draft Amnesty Week draft. It may not be polished, up to my usual standards, fully thought through, or fully fact-checked. 

Commenting and feedback guidelines:

1. This draft lacks the polish of a full post, but the content is almost there. The kind of constructive feedback you would normally put on a Forum post is very welcome.
2. This is a Forum post that I wouldn't have posted without the nudge of Draft Amnesty Week. Fire away! (But be nice, as usual)

Thank you to @Jian Xin Lim 🔸 for encouraging me to post this as part of Draft Amnesty Week.

Introduction

So, today I heard about the Consumer Reports study to do with lead in Huel (and other drinks). And I was pretty concerned about this as I drink it regularly, so spent most of today reading and discussing this. However I am a full time student, co-chair of [...]

---

Outline:

(00:17) Introduction

(00:50) Safety thresholds of heavy metals

(01:38) Huel tests and stats vs CR

(01:43) Consumer Reports

(03:33) Huel

(04:31) Chocolate

(05:09) Lead levels in Fruit and Veg

(07:05) Conclusion

(07:27) Further conclusions:

---

First published:

October 15th, 2025

Source:

https://forum.effectivealtruism.org/posts/HLbATPd5GcQSEkndG/heavy-metals-in-huel

---

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 Draft Amnesty Week draft. It may not be polished, up to my usual standards, fully thought through, or fully fact-checked. 

Commenting and feedback guidelines:

1. This draft lacks the polish of a full post, but the content is almost there. The kind of constructive feedback you would normally put on a Forum post is very welcome.
2. This is a Forum post that I wouldn't have posted without the nudge of Draft Amnesty Week. Fire away! (But be nice, as usual)

Thank you to @Jian Xin Lim 🔸 for encouraging me to post this as part of Draft Amnesty Week.

Introduction

So, today I heard about the Consumer Reports study to do with lead in Huel (and other drinks). And I was pretty concerned about this as I drink it regularly, so spent most of today reading and discussing this. However I am a full time student, co-chair of EA Bath, and have various other commitments, so this is more just a collection of my findings, rather than a post with a proper conclusion? I am also posting this as part of Draft Amnesty Week and Jian Xin recommended that it would maybe be a good idea.

Safety thresholds [...]

---

Outline:

(00:17) Introduction

(00:50) Safety thresholds of heavy metals

(01:38) Huel tests and stats vs CR

(01:43) Consumer Reports

(03:33) Huel

(04:31) Chocolate

(05:09) Lead levels in Fruit and Veg

(07:05) Conclusion

(07:27) Further conclusions:

---

First published:

October 15th, 2025

Source:

https://forum.effectivealtruism.org/posts/HLbATPd5GcQSEkndG/heavy-metals-in-huel

---

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’ve used the phrase “entertainment for EAs” a bunch to describe a failure mode that I’m trying to avoid with my career. Maybe it’d be useful for other people working in meta-EA, so I’m sharing it here as a quick draft amnesty post.

There's a motivational issue in meta-work where it's easy to start treating the existing EA community as stakeholders. The real stakeholders in my work (and meta-work in general) are the ultimate beneficiaries — the minds (animal, human, digital?) that could benefit from work I help to initiate. But those beneficiaries aren’t present to me — they aren’t my friends, they don’t work in the same building as me. To keep your eyes on the real prize takes constant work.

When that work slips, you could end up working on ‘entertainment for EAs’, i.e. something which gets great feedback from EAs, but only hazily, if [...]

---

First published:

October 17th, 2025

Source:

https://forum.effectivealtruism.org/posts/AkSDhiPuvnRNbjXAf/entertainment-for-eas

---

Narrated by TYPE III AUDIO.