This weekly roundup thread is intended for all culture war posts. 'Culture war' is vaguely defined, but it basically means controversial issues that fall along set tribal lines. Arguments over culture war issues generate a lot of heat and little light, and few deeply entrenched people ever change their minds. This thread is for voicing opinions and analyzing the state of the discussion while trying to optimize for light over heat.
Optimistically, we think that engaging with people you disagree with is worth your time, and so is being nice! Pessimistically, there are many dynamics that can lead discussions on Culture War topics to become unproductive. There's a human tendency to divide along tribal lines, praising your ingroup and vilifying your outgroup - and if you think you find it easy to criticize your ingroup, then it may be that your outgroup is not who you think it is. Extremists with opposing positions can feed off each other, highlighting each other's worst points to justify their own angry rhetoric, which becomes in turn a new example of bad behavior for the other side to highlight.
We would like to avoid these negative dynamics. Accordingly, we ask that you do not use this thread for waging the Culture War. Examples of waging the Culture War:
-
Shaming.
-
Attempting to 'build consensus' or enforce ideological conformity.
-
Making sweeping generalizations to vilify a group you dislike.
-
Recruiting for a cause.
-
Posting links that could be summarized as 'Boo outgroup!' Basically, if your content is 'Can you believe what Those People did this week?' then you should either refrain from posting, or do some very patient work to contextualize and/or steel-man the relevant viewpoint.
In general, you should argue to understand, not to win. This thread is not territory to be claimed by one group or another; indeed, the aim is to have many different viewpoints represented here. Thus, we also ask that you follow some guidelines:
-
Speak plainly. Avoid sarcasm and mockery. When disagreeing with someone, state your objections explicitly.
-
Be as precise and charitable as you can. Don't paraphrase unflatteringly.
-
Don't imply that someone said something they did not say, even if you think it follows from what they said.
-
Write like everyone is reading and you want them to be included in the discussion.
On an ad hoc basis, the mods will try to compile a list of the best posts/comments from the previous week, posted in Quality Contribution threads and archived at /r/TheThread. You may nominate a comment for this list by clicking on 'report' at the bottom of the post and typing 'Actually a quality contribution' as the report reason.
Jump in the discussion.
No email address required.
Notes -
America is a big country, inhabited by few Americans (in most of its territory). Datacenters are, all told, pretty small things, these aren't wheat fields or even airports. The politics of early singularity is interesting: people refuse the maximalist AGI and human replacement framing for assorted cultural reasons, and instead latch on to any adjacent smear – from the inane water use stuff to semi-legitimate surveillance and energy costs concerns to "they don't create jobs". The outcome is overdetermined. AI is already the main engine of economic growth, and the success at AI development and broad diffusion (at this level of utility, just scaling inference capacity) is deemed a matter of national survival. The popular resistance will be overcome, persuaded to relent, bribed in select locales, or routed around. Maybe Anthropic will finetune a Mythos 2 version specifically for propaganda.
The biggest practical consequence may be the increase in returns to Musk's idea with orbital compute, and of course his ownership of SpaceX. There are no NIMBYs in space. Without all this, it'd have taken maybe 2-3 more years until viability; now, it starts working pretty much after they manage the first Starship land-and-reuse after an orbital insertion.
This is probably because you or they won't see AGI or human replacement in your lifetimes, but they will see higher water bills, a surveillance state, higher electricity bills, and job loss. I am not sure what makes you so blasé about predicting bizarre outcomes like «singularity», but it is really odd and out-there, especially given nobody has ever predicted anything properly in the AI space. The systems are too random. Nobody knew GPT 3 would come out and expand in 2022, particularly not you. Nobody knew Opus 5 would be doing what it's doing now in 2026, you have literally no idea what 2030 or 2034 will look like. None.
I don't know why you assume "particularly not you" would have predicted GPT-3. GPT-3 follows from GPT-2 follows from Vaswani et al. Why do you think plenty of intelligent people, including Dario Amodei and Ilya Sutskever, got so agitated when that paper came out and started scrambling for commercialization? Do you believe they were that into machine translation as a business? Of course folks like Hinton, Sutskever's teacher, were running multi-decade research programs premised on connectionism being enough for intelligence in general, not just for some particular capability demo. I was pretty AGI-pilled all of my conscious life, admittedly mostly for shallow sci-fi reasons, and got convinced of inevitable singularity once I saw (and later launched) DeepDream. That degree of quasi-creative flexibility was an unambiguous proof, for me, that we have really grasped the sufficient basic primitive of universal learning. That's the biggest piece that biological life needed to go from insects to humans, and hardware&software cycles are inherently absurdly faster plus capital can grow exponentially, so what exactly could the counterargument even be?
In my opinion, this is "out there" on purely rhetorical grounds – skepticism is a product of incomplete information available in the popular discourse, and skeptics who profess to have some principled objections are continuously humbled and forced into retreat (like Chollet or LeCun, the French are annoying in this way; or hey, remember when Hlynka tried to burst my/others bubble here with some prose about mathematical Truth versus wordcel nonsense? I wonder how he feels about the recent math results from OAI/Ant). The coarse-grained logic is settled since Samuel Butler, the timelines and mechanisms were getting constantly refined, now we have a good idea of the necessary mechanisms and a thicket of more or less cheap ways to compensate for any particular – and still temporary – technological block.
Vernor Vinge, 1993:
Shane Legg, 2009:
Shane Legg, 2011:
Kurzweil, according to Google's AI, “predicts that Artificial General Intelligence (AGI) will arrive by 2029. He first made this prediction in his 1999 book The Age of Spiritual Machines, and has maintained it through subsequent books like The Singularity Is Near and The Singularity Is Nearer.”
and in 1988, “pioneering Carnegie Mellon University roboticist Hans Moravec predicted that hardware matching the computing power of the human brain would arrive by the late 2020s, enabling human-level artificial intelligence. He argued that processing capacity, driven by hardware scaling, would naturally unlock general machine intelligence.’’
While nowhere close to these giants, largely on account of my age, I believe I have a pretty decent prediction track record in this field (I have plenty of receipts here, such as talking about DeepSeek when their V1 Coder and LLM came out in late 2023; since then their architecture and much else became the default paradigm in the industry).
What about Opus 5? It's been programmed since o1-preview at least. Of course the specific product name, timing, costs, benchmark numbers and downstream capability priorities are all unpredictable but trivial. Anthropic surviving and remaining well-resourced enough to compete was very likely, the substance of the rest follows necessarily.
That's fair, that's the predictable part, as Vinge has predicted. But then again, as Yud said, you don't know how specifically a superhuman AI will beat you in chess; you can just safely bet on it doing so. I have a pretty clear idea that at this rate we will have AI doing wildly superhuman knowledge work by 2030. This is the most salient and important part, unlike the specious water use trolling. And thus the real problem people should have with AI, I believe, is still the same I outlined 3 years ago.
I strongly believe there are no remaining valid objections to this big picture. You're welcome to make some.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link