Apply to be a Teacher for TARA Round 2, 2026 🧑🏫
• Part-time role ($80 AUD/hr, ~13 hours/week).
• Lead Saturday sessions and provide remote support.
• Must have strong ML skills and completed most or all of the ARENA curriculum.
• Applications close 8 August 2026, reviewing candidates on rolling basis.
• Check out the role description and apply.
Mental health is one of the most widespread, impactable, yet neglected global health issues of our time. Even knowing mental health conditions are generally under-reported and/ or misrepresented, evidence still shows widespread effects of mental wellbeing on physical health, life satisfaction, and productivity, however it has still not taken off as a major cause area in many EA organizations I have talked to. Why is this?
I hastily vibecoded a ~live-updating EA Forum (posts + comments; I think users are static) database + semantic search because software is ~free now. I haven't rigorously checked anything over because it was/is just a random side project so use at your own risk and discretion. Enjoy!
Full database is ~17GB
In 2023-2024, it seemed a very strong bias from the EA community and the core AI safety funders that Anthropic not to be argued against or funding diverted to anything that was critical of them (directly or indirectly).
I wonder how the community and funders have updated their beliefs and behaviours?
I'm not involved in animal welfare, so I won't comment on whether broadening the tent and bringing in meat-eaters is a good thing. That said, if we take that as a given, I still find this solution of tabooing "vegan" rather strange. One direct comparison we could make here would be the 10% pledge. It seems like the community orients to the 10% pledge in much the same way you'd like people to orient towards veganism? It's a large commitment to your values. It's to be celebrated. People talk about it openly. But you aren't expected to have to sign the pledge in order to identify as EA, attend a meeting, post on the Forum, etc.
I would say it would be a very large mistake to taboo the 10% pledge entirely, and avoid mentioning it in polite company. There's a lot of gap between "Everyone must sign the 10% pledge, or you're not a real EA and should feel unwelcome here" and "Nobody should ever mention the 10% pledge in public, for fear of driving away newcomers". EA already seems to handle this well with how the 10% pledge is used. Why shouldn't that be the model for how to handle other large commitments that should be celebrated but not demanded?
Just as some feedback, I don’t think the usual objection to working on longtermism or GCRs is ‘it’s not valuable enough’, even from a general audience. I think it would be more persuasive to explain somewhere why working on these causes would achieve anything at all. It’s not clear from your post that there would be a relationship between effort and rewards, and I suspect a lot of people share the intuition that any counterfactually valuable work on the very long term future would be washed out or easily reversible.
Re Luna amendment, there was an additional Luna amendment that was trying to remove the SOB provision that died (even though the pesticide one went through). That was what I was referring to. https://www.nationalhogfarmer.com/farming-business-management/amendment-to-strike-down-save-our-bacon-act-rejected
You're right about 2, fixed. I didn't see the comment pointing this out as I don't check comments on posts more than a day after posting (otherwise I'd have way too many to check as I get a lot of comments).
Re 3, Didn’t mean to use the word PAC in the comment, was an error (not sure if I was confused at time or just a typo.). But the ampa themselves—not the PAC—are the things the donation link was to, and they’re just defeating sob. Other stuff the PAC is doing is also to defeat SOB (Craig is a big opponent of it).
I’ve been thinking about AI Safety for roughly 5 months. The more I think deeply about it, how things may play out, possible outcomes, drivers of progress etc etc, the more I realize how bad I am at thinking about it. What’s dangerous is, it’s never immediately obvious that a scenario in my mind is actually not logical. There’s so many variables, factors to consider, new knowledge you consume, new things happening in the AI world that drastically shift my view. I think humans have a tendency...
This is a linkpost for Subjective Probabilities should be Sharp by Adam Elga, which was originally published in Philosophers' Imprint in May 2010. Here is an errata for it. Below is a summary...
One could argue that 1 and 2 remain incomparable and that I have no reason to favor 2 over 1.
If the absolute value of the expected cost-effectiveness of 1 was astronomically larger than that of intervention 2, I think comparing the interventions would be similar to comparing intervention 1 with one with cost-effectiveness of 0 (burning money). It is very unclear whether the expected cost-effectiveness of 1 is positive or negative. So it would be close to arbitrary which intervention has the highest expected cost-effectiveness.
Another thing, assuming there is no 2-like intervention, is that the criterion to pick could be something other than "act straightforwardly as if you were endorsing SHARP". It could instead be some (other) form of bracketing.
Bracketing departs from impartiality, and I find this very unappealing.
Hot take: When Anthropic IPO money starts flowing into AI safety, the ecosystem should consider operating more outside charitable structures. Charities get an indirect public subsidy via tax-deductible donations, but the trade-off is greater compliance burden and cost, restrictions on spending, and a greater reliance on public goodwill. As the world gets weirder from AI-driven job losses and political shifts, taxpayers may be less happy to see their foregone tax dollars going to China dialogues or high salaries for technical AI safety researchers. (Not legal or financial advice.)
This is a condensed version of the essay linked above.
Not long ago, it was fashionable to ask “how would the AI labs ever turn a profit?” In theory, ever-increasing training costs would swallow all revenue, as labs compete to stay at the frontier. If a lab stopped training better models, their competitors would catch up, undercut them and take their market share. A lose-lose.
A...
“How long have you been v*g*n?”
This is one of the most common icebreakers at animal protection events. It’s a baseline assumption, and it mostly holds true: if you’re out advocating for animals not to be tortured or abused, realistically these days you are v**n, or close. And it makes for good conversation. It seems fairly safe to assume when you meet strangers.
But this assumption is hurting the movement in a way which we don’t always notice: someone new comes into the sp...
I strongly agree with this. Veganism should remain the gold standard. Not everyone has to reach that bar to help animals, but credit where it's due.
Thanks for raising this. I should clarify that the generalisation I've made largely applies outside of EA spaces, and the reverse tends to be true in EA spaces, where I find most people, vegan or otherwise, take the suffering of farmed animals pretty seriously.
I agree this is a challenge in EA, and I'm sure its gone well and poorly at different points in time. FWIW my experience over the last few years has been pretty positive where I have felt supported by friends and collaborators who are not vegan in my work as an animal advocate and personally as a vegan. Compared to other spaces I think overall we handle what is a tough dynamic fairly well.
I'm unsure about specifics (maybe "curse word" is too much), but I like the general direction of thought presented here. It reminded me of Matt Ball's "The End of Veganism": https://mattball.substack.com/p/the-end-of-veganism
Cross-posted to the EA Forum and LessWrong.
I've spent the past several months building and testing an applied framework — called Aster — for governing persistent human-AI relational continuity across multiple LLM platforms. My central claim is modest: some governance requirements for persistent human-AI relationships follow from observable human and institutional risks, not from any conclusion about AI sentience, and can therefore be built and t...
René Itah and I are building a public platform to track incidents of AI-caused harm to non-human animals. Some examples of these harms include autonomous vehicles hitting animals on roads, smart home systems creating unsafe conditions for companion animals, and robots interfering with nesting and migratory patterns in wildlife.
Our design ingests data from multiple different sources and uses an LLM to detect inst...
This is cross posted from X and my personal blog. Benchmarks are important to AI safety and that's why I believe METR is one of the most important safety orgs. But most benchmarks suck, and building great ones requires nuanced understanding. This post is a summary of what I've learned after 18 months focusing on benchmarks.
Pinned
Ivan Bercovich | @neversupervised
Good Benchm...
This is a crosspost from the new Animal Welfare Alignment Newsletter by Anima International. You can subscribe on Substack if you are interested in following these efforts. Audio reading also available on Substack.
The goals of this post are to:
I believe it's principally strategy 2, as well as more direct philosophical inquiries into the breadth of goals relevant here. For example, to some degree the concept of naturalness emerges in most people's image of a properly managed natural world, but naturalness is notoriously slippery and subjective, and the form of naturalness worth striving for is the sort of thing that could divide a room of competent animal ethicists.
Is it worth modifying the nervous systems of animals if this step can nearly eliminate their suffering, regardless of the lives we afford them? To what degree does the value of preserving a species diminish as we modify it from its natural state? Does conservation of species trump welfarism in any case (preserving screwworms for no positive ecological or welfare purpose)? Does population matter? Like, if we could maintain the genetic diversity of a species at 10k individuals, does maintaining a population above that level become justified when the average wellbeing level of those animals surpasses some arbitrated value? Is minimized or maximized population always better when other factors remain the same (species viability, average wellbeing)?
Is a species' reliance upon humans for survival a bad thing? Is there value in a self-maintaining ecological system as compared to one that is intensively managed? What amount of wellbeing or extinction-risk protection should be lost to gain some amount of self-maintainedness? This is basically just an elaboration of a few of the questions that arise only from the factors of naturalness, preservation, and welfarism. The necessary questions to be asked are emergent with additional factors of value. It turns into a messy field of ontological lines and consequentialist gradients as you add more.
Should we limit ourselves to management practices that operate through sterilization rather than killing? When does human activity supersede the interests of animals? What disarray could even this muddled bunch of questions be reduced to if an effort was made to parse the preferences and considerations of animals themselves, and co-balance all of these, rather than just apply anthropocentric priorities to them? There could be some level of conceptual self-awareness in some animals, and their sense of what matters about themselves could be alien to our own.
Hot take: When Anthropic IPO money starts flowing into AI safety, the ecosystem should consider operating more outside charitable structures. Charities get an indirect public subsidy via tax-deductible donations, but the trade-off is greater compliance burden and cost, restrictions on spending, and a greater reliance on public goodwill. As the world gets weirder from AI-driven job losses and political shifts, taxpayers may be less happy to see their foregone tax dollars going to China dialogues or high salaries for technical AI safety researchers. (Not legal or financial advice.)