Good point. Maybe 💀? Or maybe that should be 🫠 and aspartame should get another one...

A set of options curated from deepseek+kimi:

🙇 — fully appreciate it, but so complete there's nothing left to add

💬 — great, and I actually have something to say

😅 — can't find the right words, so here's a mildly risky joke

🤏 — agree almost entirely, just a couple of tiny hedges

🎖️ — total respect and gratitude, even though I disagree with everything

🤨 — it's all correct, but I sense a disagreement I can't place

🫥 — well-meaning, but our worldviews just don't touch

🙌 — you said what I meant, but better

🎯 — yes, that was the point of the post though

🔗 — fair point, and here's a pointer to what I already wrote (no offense)

🙏 — you caught a real mistake, thank you

😒 — right, but did it have to be that condescending?

🤦 — the confusion is on me, I wrote that badly

🗑️ — mean-spirited and useless

🌹 — I'll engage with the substance, but please remember I'm fragile

🫠 — "aspartame..." I literally cannot help myself

🤖 — are you an AI agent?

🌱 — promote away, we need more bloggers linking to themselves

Regarding the berries, if there was just a gene that said, "don't eat as much if food is running low", the individuals that had that gene would simply be outcompeted by those that don't. (Even though this is worse for everyone than if everyone slowed down.) In order for such a gene to spread, you'd need some kind of enforcement mechanism where individuals who defect are punished. That is to say, collectively beneficial behavior can emerge, but it must be fully reproductive to behavior that is selfish at the level of the gene.

I think altruism is similar, in that it can also be reduced the behavior that's selfish at the level of gene, as you say.

So, overall, I think "maximize number of offspring" is definitely not quite right. The answer is more, "maximize number of copies of genes that persist". (But even that is a little complicated as you can have gene/gene competition as in transposons!)

I want to believe. (In all seriousness, I never imagined a future where AI could be so incredibly good at coding but not at writing.) Sadly though, my best guess is that this mostly reflects the level of investment by AI companies and AI is likely to make some dramatic leaps in writing quality sometime relatively soon.

Even better, eat lots of plants!

Fixed!

I wonder who that was. Maybe Heaviside?

I'm not sure (literally not sure!) if the Postdam Declaration meets the spirit of the Franck Report and Szilard Petition. One aspect is this:

If such public announcement gave assurance to the Japanese that they could look forward to a life devoted to peaceful pursuit in their homeland and if Japan still refused to surrender, our nation might then, in certain circumstances, find itself forced to resort to the use of atomic bombs.

One the one hand, the Potsdam Declaration asks for unconditional surrender. But I don't think that's disqualifying on its own and it also includes passages like this, which I think you could reasonably argue do meet the criteria laid out above:

the Japanese military forces, after being completely disarmed, shall be permitted to return to their homes with the opportunity to lead peaceful and productive lives

Japan shall be permitted to maintain such industries as will sustain her economy and permit the exaction of just reparations in kind, but not those which would enable her to rearm for war. To this end, access to, as distinguished from control of, raw materials shall be permitted. Eventual Japanese participation in world trade relations shall be permitted.

But there's another part of both the Franck report and the Szilard Petition: They were concerned that once nuclear weapons were used, it was inevitable that other nations would develop them, e.g. this part:

If after the war a situation is allowed to develop in the world which permits rival powers to be in uncontrolled possession of these new means of destruction, the cities of the United States as well as the cities of other nations will be in continuous danger of sudden annihilation.

Though I suppose they only urge that Truman to consider these issues, and maybe he did.

I support you. Personally, after spending way too much time thinking about Bourdieu etc., I've come to believe that:

(1) The majority of worthwhile human activities have some dimension of "showing off" to them. (2) Worrying too much about showing off is best thought of as a form of neuroticism.

I mean: Marrying someone you don't like because you think they'll impress your friends or spending all your time trying to look good on Instagram is surely bad. But having a personal library is not like that!

This advice generally makes me sad, but still worth thinking about.

Ah, so the argument is more general than "reproduction" through running different physical copies, but also includes the AI self-improving? This again seems plausible to me, but still seems like something not everyone would agree with. It's possible, for example, that the "300 IQ AI" only appears at the end of some long process of recursive self-improvement, at which stage physical limits mean it can't get much better without new hardware requiring some kind of human intervention.

I guess my goal is not to lay out the most likely scenario for AI-risk, but rather the scenario that requires the fewest assumptions, that's the hardest to dispute?

I agree with you! There are a lot of things that present non-zero existential risk. I think that my argument is fine as an intellectual exercise, but if you want to use it to advocate for particular policies then you need to make a comparative risk vs. reward assessment just as you say.

Personally, I think the risk is quite large, and enough to justify a significant expenditure of resources. (Although I'm not quite sure how to use those resources to reduce risk...) But this definitely is not implied by the minimal argument.