EA Forum Podcast (All audio)

I've been pushing a moral dilemma to one side in my mind for a few months now. After Killing the ants and Killing the moths, here's the third installment in the unofficial "invertebrate trilogy."

One of the bedrooms in my flat keeps getting wasps in. I think they're crawling through gaps in our old English windows.

They arrive low-energy. Some fly around briefly, but mostly they just crawl up and down the glass (or along the floor) and slowly die. Sometimes I gather them up in a cup and release them outside. But given it's getting colder, I'm not sure that's much better. Is it better to slowly die in my flat or out in the cold?

We also had mice recently. For them, we found a humane trap and released them outside [1].

Wasps are scary too.

It's easy to have sympathy for them when they're crawling around slowly. But when the higher-energy ones get in, it's easy to lose that sympathy (which then makes me feel bad).

It's also easy to worry about how much of my attention this is taking up. I could be helping way more invertebrates by focusing on my job. But instead I'm [...]

---

First published:

January 26th, 2026

Source:

https://forum.effectivealtruism.org/posts/xYcfpm5wmCdhYkzpM/killing-the-wasps

---

Narrated by TYPE III AUDIO.

Summary

We’re a funding circle looking to give microgrants to promising individuals (and very small organisations) focused on farm animal advocacy. The microgrants will be around £500 to £2500 ($670 to $3300). We’re more likely to fund smaller requests, but we’ll consider exceptional opportunities up to £5000 ($6700). Applicants should expect to hear back from us by the end of March.

In total, we expect to give between £15,000 ($20,000) and £43,000 ($59,000).[1]

Please complete our Request for Proposals here by 1 March 2026 (end of day, anywhere in the world).

What will we fund?

We’re excited about projects to help the farm animal advocacy movement in a variety of ways. Primarily, we are looking to support individuals who want to devote their career to improving the lives of farmed animals (e.g., upskilling current advocates, or helping people transition into movement work). We’re also interested in small-scale opportunities that can directly improve the lives of animals or provide information value (e.g., an intervention pilot).

Our funding circle will consider projects based anywhere in the world. We would especially like to support work in developing countries where industrial animal farming is an emerging problem.

In general, our funding circle is [...]

---

Outline:

(00:12) Summary

(01:02) What will we fund?

(02:46) Who is eligible?

(03:18) Who is the funding circle?

(03:39) Where can I apply?

(04:09) Appendix: Longer list of potential projects

---

First published:

January 26th, 2026

Source:

https://forum.effectivealtruism.org/posts/fP9hvx8A9MG9F4rEb/animal-microgrants-request-for-proposals

---

Narrated by TYPE III AUDIO.

This is a link post.

We're giving away 100 Aerolamp DevKits, a lamp that kills germs with far-UVC.

Are you sick of getting sick in your group house? Want to test out fancy new tech that may revolutionize air safety?

Claim your Aerolamp

What is far-UVC?

Far-UVC is a specific wavelength of ultraviolet light that kills germs, while being safe to shine on human skin. You may have heard of UV disinfection, used in eg hospitals and water treatment. Unfortunately, conventional UVC light can also cause skin and eye damage, which is why it's not more widely deployed.

Far-UVC refers to a subset of UVC in the 200-235 nm spectrum, which has been shown to be safe for human use. Efficacy varies by lamp and setup, but Aerolamp cofounder Vivian Belenky estimates they may be "roughly twice as cost effective on a $/CFM basis", compared to a standard air purifier in a 250 square foot room.

For more info, check out faruvc.org, or the Wikipedia page on far-UVC.

Why are we giving away lamps?

Far-UVC deserves to be everywhere. It's safe, effective, and (relatively) cheap; we could blanket entire cities with lamps to drive down seasonal flu, or prevent the next [...]

---

Outline:

(00:28) What is far-UVC?

(01:23) Why are we giving away lamps?

(02:52) Who can get one?

(03:30) Whos behind this?

---

First published:

January 26th, 2026

Source:

https://forum.effectivealtruism.org/posts/duwzFZYuyciazWzHG/aerodrop-far-uvc-lamp-giveaway

Linkpost URL:
https://aerodrop.org/

---

Narrated by TYPE III AUDIO.

I think accelerating AI is justified largely because of the enormously positive impacts I expect it to bring to billions of people alive today: faster medical developments, dramatically increased economic prosperity, and much greater product variety from technological innovation.

Some have asked me: why do I focus on these short-term impacts of accelerating AI to justify my pro-acceleration view, rather than on the impacts that might occur over the upcoming billions of years?

The first thing worth pointing out about this question is that, at least in my experience, it almost never comes up outside of debates about AI. In virtually every other area of human decision-making, people generally accept without much argument that the very-long-term consequences of our actions are extremely difficult to predict. Because of this extreme uncertainty, most people intuitively recognize that we should place less weight on projected very-long-term effects, since we simply cannot know with any confidence what those effects will actually turn out to be.

When policymakers debate housing regulations, or when patients decide whether to undergo a medical procedure, or when nations consider going to war, almost nobody would accept an argument structured like this: "We should willingly accept devastating costs over [...]

---

First published:

January 26th, 2026

Source:

https://forum.effectivealtruism.org/posts/oFsEsECrHzW6KBXJW/why-i-reject-the-longtermist-argument-against-accelerating

---

Narrated by TYPE III AUDIO.

This is a link post.

This is a rough research note – we’re sharing it for feedback and to spark discussion. We’re less confident in its methods and conclusions.

Summary

Different strategies make sense if timelines to AGI are short than if they are long.

In deciding when to spend resources to make AI go better, we should consider both:

1. The probability of each AI timelines scenario.
2. The expected impact, given some strategy, conditional on that timelines scenario.

We’ll call the second component "leverage." In this note, we'll focus on estimating the differences in leverage between different timeline scenarios and leave the question of their relative likelihood aside.

People sometimes argue that very short timelines are higher leverage because:

1. They are more neglected.
2. AI takeover risk is higher given short timelines.

These are important points, but the argument misses some major countervailing considerations. Longer timelines:

1. Allow us to grow our resources more before the critical period.
2. Give us more time to improve our strategic and conceptual understanding.

There's a third consideration we think has been neglected: the expected value of the future conditional on reducing AI takeover risk under different timeline scenarios. Two factors pull in [...]

---

Outline:

(00:24) Summary

(02:29) Timelines scenarios and why they're action-relevant

(07:50) Understanding leverage

(11:09) Takeover impact

(11:20) The default value of the future is higher on medium timelines than short or long timelines

(17:54) Shorter timelines allow for larger AI takeover risk reduction

(20:53) Whether short or medium timelines are highest leverage depends on the resources or skillset being deployed

(22:35) Trajectory impact

(22:52) The default probability of averting AI takeover is higher on medium and longer timelines

(25:12) It's unclear whether feasible value increase is greater or lower on different timelines

(27:15) Conclusion

(29:08) Appendix: BOTEC estimating the default value of the future on different timelines

---

First published:

January 22nd, 2026

Source:

https://forum.effectivealtruism.org/posts/zANckKgyE4tDkynbG/are-short-ai-timelines-really-higher-leverage

Linkpost URL:
https://www.forethought.org/research/short-timelines-arent-obviously-higher-leverage

---

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 thought it'd be cool if EA were to have its own version of Inkhaven or Halfhaven. I am writing this to gauge interest!

Basics: it would be online-only, on a Discord server and take place from March 1-31. Every day, you write a blog post of at least 500 words by your timezone's midnight. You'll be in a small sub-group who provide feedback and keep each other accountable to posting, with optional coworking calls.

Sign up here. If we hit 10 people, we'll go through with it.

LOGISTICS

Let's port what was done well from Halfhaven and any learning on how to do it better. I like the simple post-a-day intensity of Inkhaven.

• When: March 1–31, 2026
• Where: A Discord server (link provided after we hit 10 participants), with coworking
• Commitment: Publish one 500+ word post per day by your timezone's midnight
• Where to publish: Your personal blog, Substack, EA Forum, LessWrong; anywhere that's publicly accessible.

I'm kickstarting this with a pact. A pact is an agreement to do a thing if and only if enough people commit to doing it (also known as an assurance contract). Lowkey, this is a soft launch for my platform [...]

---

First published:

January 25th, 2026

Source:

https://forum.effectivealtruism.org/posts/GX4aN6TMYCvbmNHpa/proposing-inkwell-inkhaven-for-eas

---

Narrated by TYPE III AUDIO.

This is a link post.

At this point, I am confident that philanthropy is the best area for me to work in, and I have a clearer sense of what I plan to do in the area.

Broad Area: Philanthropy

Over time, I have become increasingly convinced that philanthropy is the right area for me. I think this applies both broadly to many people considering impact and more specifically to me from a personal fit perspective.

Broadly, I find the area exciting for a few reasons. Much like starting a charity incubator (CE, my most recent project), I think the market is both large and non-competitive. Most philanthropic ecosystems and meta-organisations evolved fairly organically, and as such there are many gaps and opportunities for improvement. The effective giving ecosystem is both large and growing, and I think it is common for a philanthropic advisor to be able to influence significantly more funds than they could earn to give, particularly if their strengths are in communication.

From a personal fit perspective, there are substantial opportunities for impact, and my background and connections make me well placed to make a big difference. Despite not even having publicly launched, I have back-to-back monthly [...]

---

First published:

January 25th, 2026

Source:

https://forum.effectivealtruism.org/posts/ufTaYjRxwEWktdGAs/my-career-plan-launching-elevate-philanthropy

Linkpost URL:
My Career Plan: Launching Elevate Philanthropy!

---

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.

TLDR: Will AI-automation first speed up capabilities or safety research? I forecast that most areas of capabilities research will see a 10x speedup before safety research. This is primarily because capabilities research has clearer feedback signals and relies more on engineering than on novel insights. To change this, researchers should now build and adopt tools to automate AI Safety research, focus on creating benchmarks, model organisms, and research proposals, and companies should grant differential access to safety research.

Epistemic status: I spent ~ a week thinking about this. My conclusions rely on a model with high uncertainty, so please take them lightly. I’d love for people to share their own estimates about this.

The Ordering of Automation Matters

AI might automate AI R&D in the next decade and thus lead to large increases in AI progress. This is extremely important for both the risks (eg an Intelligence Explosion) and the solutions (automated alignment research).

On the one hand, AI could drive AI capabilities progress. This is a stated goal of Frontier AI Companies (eg OpenAI aims for a true automated AI researcher by March of 2028), and it is central to the AI 2027 forecast. On the other hand [...]

---

Outline:

(01:00) The Ordering of Automation Matters

(05:42) Methodology

(11:46) Prediction

(14:00) Levers for affecting the order

(14:53) Speeding up Safety Automation

(19:27) Future Work: How could this be researched properly?

---

First published:

January 22nd, 2026

Source:

https://forum.effectivealtruism.org/posts/mNJ8f3sZrquFvYXgi/do-we-get-automated-alignment-research-before-an-ai-takeoff

---

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.

Summary

In January 2025, FarmKind ran a provocative media campaign which used controversial media messaging and materials to promote ‘offsetting’ as an option for individuals who are concerned about factory farming but are currently unwilling or unable to change their diet.

The campaign raised an estimated $16,700--$59,300 (explained in our Results section below) and generated a number of media ‘hits’ including TV and created some debate that many advocates have told us they found productive. However we made mistakes in its execution and generated unproductive controversy within the EA and animal advocacy movements.

This post aims to explain our theory of change, what happened, what we got wrong, and what we learned.

We still believe mobilizing the meat-eating majority to take action for farmed animals requires meeting them where they're at, which sometimes means provocative framing that distinguishes us from vegan advocacy -- though we understand many in the movement disagree. However, we regret specific execution failures, particularly our insufficient stakeholder consultation, which risks sparking infighting within the animal movement.

Context

FarmKind is a donation platform that aims to bring more money into the movement against factory farming. People donate through our platform directly to six highly effective farmed [...]

---

Outline:

(00:12) Summary

(01:23) Context

(02:06) The goals of our campaign

(02:51) Primary goals

(03:16) Secondary goals

(04:06) How we envisaged it working

(05:24) Launching the campaign

(07:18) Coordination with Veganuary

(07:22) Did you tell Veganuary about the campaign in advance?

(08:21) Did Veganuary object to the campaign?

(09:07) Is there bad blood between you and Veganuary?

(09:50) Does Veganuary endorse this campaign?

(10:13) What we got wrong

(10:16) 1) Underestimating the risk of movement infighting

(12:29) 2) Insufficient stakeholder consultation

(13:11) 3) Internal coordination failures

(13:52) How we responded to concerns

(17:20) Results

(20:12) FAQs

(20:15) Are you anti-vegan?

(22:35) Aren't you concerned about dissuading people from being vegan?

(26:07) Have you measured whether you're dissuading people from being vegan or supporting animal advocacy?

(29:03) Why not just do something much more nuanced?

(29:45) Why did you pitch to tabloids and right-wing outlets?

(30:50) Conclusion

---

First published:

January 23rd, 2026

Source:

https://forum.effectivealtruism.org/posts/c2buSr3oatKQJZi6F/reflections-on-farmkind-s-january-media-campaign

---

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.

Introduction to the Digital Consciousness Model (DCM)

Artificially intelligent systems, especially large language models (LLMs) used by almost 50% of the adult US population, have become remarkably sophisticated. They hold conversations, write essays, and seem to understand context in ways that surprise even their creators. This raises a crucial question: Are we creating systems that are conscious?

The Digital Consciousness Model (DCM) is a first attempt to assess the evidence for consciousness in AI systems in a systematic, probabilistic way. It provides a shared framework for comparing different AIs and biological organisms, and for tracking how the evidence changes over time as AI develops. Instead of adopting a single theory of consciousness, it incorporates a range of leading theories and perspectives—acknowledging that experts disagree fundamentally about what consciousness is and what conditions are necessary for it.

Here, we present some of the key initial results of the DCM. The full report is now available here.

We will be hosting a webinar on February 10 to discuss our findings and answer audience questions. You can find more information and register for that event here.

Why this matters

It is important to assess whether AI systems might be conscious in a [...]

---

Outline:

(00:13) Introduction to the Digital Consciousness Model (DCM)

(01:31) Why this matters

(02:51) Why estimating consciousness is a challenging task

(03:38) How the model works

(05:38) How to interpret the results

(07:57) Key findings

(10:18) What's next

(10:56) Acknowledgments

---

First published:

January 23rd, 2026

Source:

https://forum.effectivealtruism.org/posts/ojLqrYgrEdqWNgb8H/digital-consciousness-model-results-and-key-takeaways

---

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.