As an American, I think this is actually a really good point. The only reason to even consider racing with China to build AI is if we believe the outcome is better if America “wins”. Seems worth keeping an eye on the political situation in the US, e.g. for stuff like PEPFAR cancellation, to check if the US “winning” is actually better from a humanitarian perspective. Hopefully lab leaders are smart enough to realize this.
Lab leaders are probably trying mostly to maximize the value of their company, not the value of the world in my view. (Doesn’t mean they give zero weight to moral considerations.) Also, if the US government realizes that they are reasoning along the lines of “let’s slow down development because it doesn’t matter if the US beats China”, the US government will probably find ways to stop them being lab leaders.
I wasn’t really reading anything in this post as favoring the US over China in an “us vs them” way, so much as I was reading a general anxiety about multipolar scenarios. Also, very few of us are based in China relative to the US or EU. ‘The country with the community of people who have been thinking about this the longest’ is probably the country you want the thing to happen in, if it has to happen (somewhere).
fwiw I’m usually very anxious about this us vs them narrative, and didn’t really feel this post was playing into it very strongly, other than in the above ways, which are (to me) reasonable.
The author is using “we” in several places and maybe not consistently. Sometimes “we” seems to be them and the readers, or them and the EA community, and sometimes it seems to be “the US”. Now you are also using an “us” without it being clear (at least to me) who that refers to.
Who do you mean by ‘The country with the community of people who have been thinking about this the longest’ and what is your positive evidence for the claim that other communities (e.g., certain national intelligence communities) haven’t thought about that for at least as long?
I took your comment to have two parts it was critiquing:
The floating use of the word ‘we’ (which I agree is ambiguous, but basically don’t care about).
The ‘us vs China’ frame as a whole (which has some surface area with the floating antecedent of the word ‘we’, but is a distinct and less pedantic point).
My response was addressing point 2, not point 1, and this was intentional. I will continue not to engage on point 1, because I don’t think it matters. If you’re devoted to thinking Lintz is a race-stoking war hawk due to antecedent ambiguity in a quickly-drafted post, and my bid to dissuade you of this was ineffective, that’s basically fine with me.
“Who do you mean by ‘The country with the community of people who have been thinking about this the longest’”
The US — where ~all the labs are based, where ~all the AI safety research has been written, where Chinese ML engineers have told me directly that, when it comes to AI and ML, they think exclusively in English. Where a plurality of users of this forum live, where the large models that enabled the development of DeepSeek were designed to begin with, where the world’s most powerful military resides, where....
Yes, a US-centric view is an important thing to inspect in all contexts, but I don’t think it’s unreasonable to think that the US is the most important actor in the current setting (could change but probably not), and to (for a US-based person who works on policy… in the US, as Lintz does, speaking to a largely US-based audience) use ‘we’ to refer to the US.
“What is your positive evidence for the claim that other communities (e.g., certain national intelligence communities) haven’t thought about that for at least as long?”
I want to point out that this is unfair, since meeting this burden of proof would require comprehensive privileged knowledge of the activities over the past half century of every intelligence agency on the planet. My guess is you know that I don’t have that!
Things I do know:
People who’ve been in touch with US intelligence agencies since ~2017 have reported that getting them to take AI seriously has been an uphill battle (not only from an x-risk perspective, but as ‘a thing that will happen at all’).
US intelligence agencies are currently consulting with senior folks in our sphere, at least occasionally, which indicates that they may lack confident inside-view takes on the situation.
The US is currently dropping the ball relative to its stated goals (i.e. Taiwan tariffs), which indicates that the more informed parts of government either don’t have much power or aren’t all that informed actually! (probably mostly the former, but I’m sure the latter has weight, too)
Finally: “Positive evidence.… have not” is a construction I would urge you to reconsider — positive evidence for a negative assertion is a notoriously difficult thing to get.
I’m confused by your seeming “we vs China” viewpoint—who is this “we” that you are talking about?
As an American, I think this is actually a really good point. The only reason to even consider racing with China to build AI is if we believe the outcome is better if America “wins”. Seems worth keeping an eye on the political situation in the US, e.g. for stuff like PEPFAR cancellation, to check if the US “winning” is actually better from a humanitarian perspective. Hopefully lab leaders are smart enough to realize this.
Lab leaders are probably trying mostly to maximize the value of their company, not the value of the world in my view. (Doesn’t mean they give zero weight to moral considerations.) Also, if the US government realizes that they are reasoning along the lines of “let’s slow down development because it doesn’t matter if the US beats China”, the US government will probably find ways to stop them being lab leaders.
I wasn’t really reading anything in this post as favoring the US over China in an “us vs them” way, so much as I was reading a general anxiety about multipolar scenarios. Also, very few of us are based in China relative to the US or EU. ‘The country with the community of people who have been thinking about this the longest’ is probably the country you want the thing to happen in, if it has to happen (somewhere).
fwiw I’m usually very anxious about this us vs them narrative, and didn’t really feel this post was playing into it very strongly, other than in the above ways, which are (to me) reasonable.
The author is using “we” in several places and maybe not consistently. Sometimes “we” seems to be them and the readers, or them and the EA community, and sometimes it seems to be “the US”. Now you are also using an “us” without it being clear (at least to me) who that refers to.
Who do you mean by ‘The country with the community of people who have been thinking about this the longest’ and what is your positive evidence for the claim that other communities (e.g., certain national intelligence communities) haven’t thought about that for at least as long?
I took your comment to have two parts it was critiquing:
The floating use of the word ‘we’ (which I agree is ambiguous, but basically don’t care about).
The ‘us vs China’ frame as a whole (which has some surface area with the floating antecedent of the word ‘we’, but is a distinct and less pedantic point).
My response was addressing point 2, not point 1, and this was intentional. I will continue not to engage on point 1, because I don’t think it matters. If you’re devoted to thinking Lintz is a race-stoking war hawk due to antecedent ambiguity in a quickly-drafted post, and my bid to dissuade you of this was ineffective, that’s basically fine with me.
“Who do you mean by ‘The country with the community of people who have been thinking about this the longest’”
The US — where ~all the labs are based, where ~all the AI safety research has been written, where Chinese ML engineers have told me directly that, when it comes to AI and ML, they think exclusively in English. Where a plurality of users of this forum live, where the large models that enabled the development of DeepSeek were designed to begin with, where the world’s most powerful military resides, where....
Yes, a US-centric view is an important thing to inspect in all contexts, but I don’t think it’s unreasonable to think that the US is the most important actor in the current setting (could change but probably not), and to (for a US-based person who works on policy… in the US, as Lintz does, speaking to a largely US-based audience) use ‘we’ to refer to the US.
“What is your positive evidence for the claim that other communities (e.g., certain national intelligence communities) haven’t thought about that for at least as long?”
I want to point out that this is unfair, since meeting this burden of proof would require comprehensive privileged knowledge of the activities over the past half century of every intelligence agency on the planet. My guess is you know that I don’t have that!
Things I do know:
People who’ve been in touch with US intelligence agencies since ~2017 have reported that getting them to take AI seriously has been an uphill battle (not only from an x-risk perspective, but as ‘a thing that will happen at all’).
US intelligence agencies are currently consulting with senior folks in our sphere, at least occasionally, which indicates that they may lack confident inside-view takes on the situation.
The US is currently dropping the ball relative to its stated goals (i.e. Taiwan tariffs), which indicates that the more informed parts of government either don’t have much power or aren’t all that informed actually! (probably mostly the former, but I’m sure the latter has weight, too)
Finally: “Positive evidence.… have not” is a construction I would urge you to reconsider — positive evidence for a negative assertion is a notoriously difficult thing to get.