I upvoted the article, it makes good points. But personally, I will mostly continue treating insects as moderately important. Your article implicitly assumes pure utilitarianism. Utilitarian calculations play an important role in my decision-making, but I don't listen to them religiously. If I did, then there might still be more important things than insect suffering. 

For example, I once thought that the conclusion of utilitarianism is that we should try to turn everything in the universe into hedonium (a homogeneous substance with limited consciousness, which is in a constant state of supreme bliss), even if our chances of success are minuscule (I see someone else argued for it here). But then I realised that I'm just not excited about that. So I concluded that I'm not a pure utilitarian. This argument about insects also makes me feel like I'm not a pure utilitarian.

shown by the beautifully scribbled light blue area

The scribble is indeed very beautiful.

In your graph above, it looks like impact for a lot more than one year. I assume it's something like this:
 

The red line here is what would've happened without Stop The Farms campaign, and blue line shows that it's different for a little while with the campaign. But I assume that the market soon (like within a year) returns back to the same growth trajectory, and it's as if we never did anything, except that maybe farms are build in a different country. Chicken production is growing and I don't think this will change in the relevant timeframe of few years.

In general, if you are worried that animal advocacy efforts will soon become irrelevant because the world will change a lot soon, it could make sense to donate to charities that have impact quickly. Shrimp Welfare Project might qualify. But maybe it makes more sense to try to find a way to impact the welfare of animals in the post-AGI world somehow, even though it's really unclear how to do that.

Regarding the cells I28:30, yes you could do that, it would change estimates for cage-free and broiler reforms. If you think these yearly probability that commitments become irrelevant should be higher, I'd be curious for which reason. Possible reasons I listed include x-risks, global catastrophic risks, societal collapse, cultured meat taking over, animals bred not to suffer, black swans.

For context, my choices for "Yearly decrease in probability that commitment is relevant" numbers are informed by this forecast which predicts that the number of chickens slaughtered for meat will be roughly the same in 2052 as it is now, but just 12% of what it is now in 2122. My value for 2122 is slightly lower, 11% because that meticulous question also has this condition: "If humanity goes extinct or ceases to have a developed society prior to a listed year, that sub-question will resolve as Ambiguous." I only decreased the forecast for 2122 slightly because this forecast predicts that the probability of human extinction before 2100 is just 1%, although looking back at this, I think I could've adjusted for x-risks more because much higher estimates of x-risks seem reasonable.

Thank you for an interesting comment. 

I'm aware of zdgroff's analysis. In the context of my analysis, I guess it would inform how long the ban of fur farming in Poland might last. But the possibility of fur farming being banned in Poland and then the ban being lifted some years later hadn't even occurred to me. I am much more worried about production moving to other countries to meet the same demand, as this has happened before. I imagine that investors into fur farming would choose to build farms in one of the many countries that allow fur farming, rather than lobby a country like Poland to rescind its ban.

Actually, it is also relevant for a possible EU cage-free ban. I can imagine that ban being rescinded. I don't think this consideration would affect the results of my estimate much, though it does complicate thinking about how many years impact lasts a little bit.

Thanks for a very useful post. I can't find: does this account for the differences in moral weight and probability of sentience between different animals. If yes, how?

Thanks for working on this. I just want to point out that if a charity helps say one animal per dollar, the real cost for the animal advocacy movement is a bit higher if you account for the following:

• Opportunity cost of staff: People working at animal advocacy charities often accept lower salaries than they could earn elsewhere. Some might have been earning-to-give if they weren't directly working in the field, potentially donating substantial amounts to animal causes.
• Hidden costs: Volunteer time, pro bono services, and other non-monetary contributions often aren't factored into cost calculations but represent real resources.
• Diminishing returns: animal advocacy interventions may be becoming less cost-effective over time as the easiest wins ("low-hanging fruits") were addressed first.
• Failed interventions: For every successful approach we discover, there were likely multiple attempted interventions that didn't work. The "research and development" costs of finding effective strategies should ideally be factored into overall movement costs.
  
  I don't know if I advise you to change anything based on this though. Your estimates are already quite conservative and perhaps it's best to avoid complicating things with considerations like these.

Yeah, that doesn't look right. I recommend looking at the spreadsheet rather than the post. I updated some parts of it at some point last year. I see in the spreadsheet that the 5th graph now looks like this

But I don't know if that's still up to date. I haven't been following the progress lately, but many of these broiler commitments are not being implemented, unfortunately. 

Interesting points. Starting with 27:45, there are two talks here that claim that AI will probably be bad for farmed animals. @Sam Tucker urges in his talk to work on banning AI in animal farms. There is also discussion on it at 58:27 where Sam says that he is 99% sure that AI in farms will be bad for animals, if I understood him correctly, partly because it might allow factory farming to stay around for longer. Perhaps you should discuss this issue with Sam.

I understand where you're coming from but I wonder whether this would also have negative consequences. Perhaps it would increase the pace of AI development. It would make LLMs more useful, which might increase investments into AI even more. And maybe it would also make LLMs generally smarter, which could also accelerate AI progress (this is not my area, I'm just speculating). Some EA folks are protesting to pause AI, increased progress might not be great. It would help all the research, but not all research makes the world better. For example, it could benefit research into more efficient animal farming, which could be bad for animals. Considerations like these would make me too unsure about the sign of the impact to eagerly support such a cause, unfortunately.