I'd also not want it completely removed, I meant lessened. The case is that community posts are overrated by the karma system. I find that this can have the effect of making the forum feel insular and navel-gazing, and distracted from good object-level analysis. The current system counteracts that by putting them on the bottom.
I don't understand why people are down voting this so much. There needs to be far more reflection on "what if we are wrong?" I say this as someone who is fairly confident AI is not in a bubble. But so much is staked on this; reputations, grant decisions, career decisions.
The EV of grappling with the possibility is high. At the very least, it helps steer away from the asymmetric status-quo fixation on bull case proponents. It does not challenge the fundamental thesis that AI is the most consequential technology left. Things could just take a little longer then predicted!
I'd love to see an equivalent report, but for the rate of benchmark creation per subject area. It's one thing to hear that progress on AECI to be accelerating, but it's not to my mind a major update on the status quo, unless AECI itself is improving quickly as well.
Another issue is that there is no longer any org that produces career prioritization research.
I suspect people have the mistaken notion that this area of research is just a reflection of philanthropic prioritization; the places it is advisable to recommend donations are also the best to recommend for careers. But this is a major simplification, even though there's a lot of overlap. Altrusitically optimal career trajectories are waaaaay broader and more obscure due to personal fit and labour market demand considerations.
Probably good doesn't do this. 80k used to do this well, but now they are just another AI safety advocacy group (no slight intended, those are needed too).
You are indeed in the right place! I do like to see more wild animal welfare discussions happening here. Regarding the forum norms, the animal welfare and opportunities tags are most appropriate for this, but it's not a big deal. If you feel like writing a longer post outside quick takes, you may find the wild animal topic tab helpful to see what others have written.
We know that unhatched chick embryos can suffer late in development, but current evidence is mixed on mid-term development around day-13. The in-ovo sexing technology used by the NestFresh brand in the US works on day-13 (which is banned in Germany). Claude's overview is here. Given that public policy and industry tech adoption is still in early stages outside Europe, it seems to me that it is great time to fund further research on chick embryo sentience. Before day-13 becomes standard.
I think the forum team has generally shown good judgement on which posts to feature. But still, it is hard to see how this doesn't give too much authority to too few decision makers on what content does and doesn't get boosted. A major benefit of a public forum is that it can help surface new ideas by casting a wide net; but centering featured posts this much privileges established authorities.[1]
Some amount of featuring is good, particularly of old posts that can get buried (Lesswrong does this well!) or when it can concentrate the forum on one topic, as it has effectively during debate weeks. The balance is just off with the current design.
I definitely not the "worst" PR crisis. The DOW account is a random poster for the Office of the Under Secretary of War for Research and Engineering, not Hegseth.
But I agree that it would be best for Republicans not to be polarized against AI safety. They aren't so far; thanks in part to Speaker Johnson. I suggest not letting random trolls goad us into abandoning courtesy and the presumption of good will. It is counterproductive to respond to much of this stuff, as that imbues it with more seriousness than it deserves.
If we broaden this to "value change" and maintain it as a future aspiration rather than a current priority, then I'm for it.
Some limited form of value change seems necessary for progress on wild animals and digital minds, so I don't think the topic can be avoided indefinitely.
My two cents is that the movement has yet to achieve widespread value change even in highly-educated, academic circles. I think it would be highly effective on the margin to focus on think tank staffers, elite law school students, graduate students in ecology subfields, and so on. There should be a consensus understanding on these issues - for instance, that animal suffering is a negative externality of meat eating.
Just as there came to be with the claim that climate change is occuring, and is anthropogenic.
I'd also not want it completely removed, I meant lessened. The case is that community posts are overrated by the karma system. I find that this can have the effect of making the forum feel insular and navel-gazing, and distracted from good object-level analysis. The current system counteracts that by putting them on the bottom.
I think that's already the case on the main page. Or maybe you'd like a comment pop-up bubble when hovering over a link? I wouldn't mind that.
I don't understand why people are down voting this so much. There needs to be far more reflection on "what if we are wrong?" I say this as someone who is fairly confident AI is not in a bubble. But so much is staked on this; reputations, grant decisions, career decisions.
The EV of grappling with the possibility is high. At the very least, it helps steer away from the asymmetric status-quo fixation on bull case proponents. It does not challenge the fundamental thesis that AI is the most consequential technology left. Things could just take a little longer then predicted!
I'd love to see an equivalent report, but for the rate of benchmark creation per subject area. It's one thing to hear that progress on AECI to be accelerating, but it's not to my mind a major update on the status quo, unless AECI itself is improving quickly as well.
Another issue is that there is no longer any org that produces career prioritization research.
I suspect people have the mistaken notion that this area of research is just a reflection of philanthropic prioritization; the places it is advisable to recommend donations are also the best to recommend for careers. But this is a major simplification, even though there's a lot of overlap. Altrusitically optimal career trajectories are waaaaay broader and more obscure due to personal fit and labour market demand considerations.
Probably good doesn't do this. 80k used to do this well, but now they are just another AI safety advocacy group (no slight intended, those are needed too).
You are indeed in the right place! I do like to see more wild animal welfare discussions happening here. Regarding the forum norms, the animal welfare and opportunities tags are most appropriate for this, but it's not a big deal. If you feel like writing a longer post outside quick takes, you may find the wild animal topic tab helpful to see what others have written.
@Wild_Animal_Initiative may also be relevant for you: https://www.wildanimalinitiative.org/grants
Plus there's some wild animal topic conversation on the Sentient Futures slack.
We know that unhatched chick embryos can suffer late in development, but current evidence is mixed on mid-term development around day-13. The in-ovo sexing technology used by the NestFresh brand in the US works on day-13 (which is banned in Germany). Claude's overview is here.
Given that public policy and industry tech adoption is still in early stages outside Europe, it seems to me that it is great time to fund further research on chick embryo sentience. Before day-13 becomes standard.
I think the forum team has generally shown good judgement on which posts to feature. But still, it is hard to see how this doesn't give too much authority to too few decision makers on what content does and doesn't get boosted. A major benefit of a public forum is that it can help surface new ideas by casting a wide net; but centering featured posts this much privileges established authorities.[1]
Some amount of featuring is good, particularly of old posts that can get buried (Lesswrong does this well!) or when it can concentrate the forum on one topic, as it has effectively during debate weeks. The balance is just off with the current design.
It's already the case that the weighted upvote system has this effect.
I definitely not the "worst" PR crisis. The DOW account is a random poster for the Office of the Under Secretary of War for Research and Engineering, not Hegseth.
But I agree that it would be best for Republicans not to be polarized against AI safety. They aren't so far; thanks in part to Speaker Johnson. I suggest not letting random trolls goad us into abandoning courtesy and the presumption of good will. It is counterproductive to respond to much of this stuff, as that imbues it with more seriousness than it deserves.
If we broaden this to "value change" and maintain it as a future aspiration rather than a current priority, then I'm for it.
Some limited form of value change seems necessary for progress on wild animals and digital minds, so I don't think the topic can be avoided indefinitely.
My two cents is that the movement has yet to achieve widespread value change even in highly-educated, academic circles. I think it would be highly effective on the margin to focus on think tank staffers, elite law school students, graduate students in ecology subfields, and so on. There should be a consensus understanding on these issues - for instance, that animal suffering is a negative externality of meat eating.
Just as there came to be with the claim that climate change is occuring, and is anthropogenic.