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 -
Is it over for AI Safety? From an interview:
https://truthsocial.com/@realDonaldTrump/posts/117269745153543631
Meanwhile in China, Dario's call for AI pacing have not gone down well. Understandably, they are not enthusiastic about the US achieving a permanent high ground in a key technology after the last few years of semiconductor sanctions and suppression. Also they can tell that Dario doesn't like the Party at all and seems to be openly plotting its downfall.
Dario's essay: https://darioamodei.com/post/we-must-pace-the-frontier
A Chinese semi-official response: https://www.globaltimes.cn/page/202609/1370436.shtml
So if Trump is keen for an AI race and so is China, then it looks pretty much over for Safety? Even the other US tech companies have only provided lip service to slowing things down. Elon said Dario's sentiment was good but is still moving quickly. Sam Altman says all kinds of things about safety, while displaying heroic negligence in training AIs. I don't doubt the sincerity of employees who quit, they have many financial reasons to go with the flow and downplay risks. Lurid rumours of Aella promising reverse gangbangs to those who quit frontier AI companies don't neccessarily outweigh the personal financial gains. It seems silly to doubt those at the frontier who are frightened when we ourselves have no special insights. In pure observational terms they're better equipped than anyone, only in policy or analysis should we be disagreeing, I think.
I also find it comical that the AI-safety camp spent such a long time building up institutions and factions and entire AI companies but seem to have been totally snookered by Jensen's suave dinner party skills and the appeal of fun AI videos to a Boomer's Boomer plus the raw partisanship of 'Dems want it, so I don't'. Even Aschenbrenner managed to do better at Trump-whispering. Shortly after he wrote his essay, I recall a video where the fundamentals were accepted by Trump, someone told him about it. And it did fit well with his instincts, getting one up on China + more energy production.
Right now Trump is reTruthing Fox News clips framing AI safety as a global one world government Democrat control-seeking scheme: https://truthsocial.com/@realDonaldTrump/posts/117273026439561125
I don't see any AI safety people on right wing twitter either, those who acknowledge the issue at all still would prefer to let it rip than hand the left/big tech elites control of the world, favouring decentralization of AI instead.
I wonder whether it would always have gone down this way, or whether Kamala was going to do things differently? Would Jensen's marketing skills and the need to keep stocks up + compete with China prevail over the Bernie Sanders-style Left element? Probably not? The Dario vision does seem much more closely aligned with Democrats, Civil Society and so on. But on the other hand, the US government is exceptionally slow to legislate and big tech has strong lobbying power.
And who knows if it will stay this way. Sometime soon there's probably going to be a much bigger AI swarm incident, which might yet incite a hysterical kneejerk reaction.
I'm just glad that we're finally past the totally clueless people insisting AI is not even useful for coding and are on to the next stage of grief we have to drag people through on the way to acceptance that big Yud has been right the entire time. The faux cynical takes that the labs at the absolute cutting edge of this are fudding their own product to the dismay of the sitting president in some insane 8 dimensional chess move because they want to be regulated by Bernie Sanders is so plainly retarded that it can't possibly last very long.
Yud was wrong about the main thing which was the fundamentals of how AI was going to be developed. He was right about X risk, but so were a lot of people, this has been a common fear since the invention of the first complex machines, or even machines in general. His delusion about safety was more about his belief that there was any stopping this train before it reaches its destination, which there isn’t.
As far as being wrong about the fundamentals of how AI was going to be developed I think the most wrong thing he was confident about was the idea of FOOM and in general not predicting this multi-year phase we're currently in as training and scaling continue to produce steady returns in capabilities. Ironically if the tranformer architecture came out in like 2030 instead of 2017 and compute was abundant relative to today we might have seen a FOOM. Albeit I concede on a lot of the specific substrate stuff both he and Hansen talked about were well off. In their defense nearly no one else registered a prediction at all so while we can hold this against them I object to raising anyone else's predictive power relative to them.
His stuff on alignment in general all holds up pretty well and I'm not so sure you can really accuse him of believing rather than hoping that there are stops on the train. He seems to have correctly intuited that trying to add breaks was a monumental task from a remarkable distance.
Unlikely, I do think you are missing something fundamental. One of the big changes these days from the early days of yore, is that now there is money in creating and curating hard-expert level datasets. Back in the day if you wanted data to train some model, you spend money scraping the internet for data that was not explicitly designed for ML and then money labeling it. It is expensive and time consuming. This is the ImageNet stuff. These days because transformers and the like have demonstrated such a fundamental capacity, and the business folks and VCs finally see dollar signs, there is a strong business case to create data explicitly for ML research. They literally hire experts to write Question-Answer pairs with intermediary steps for model training. It costs orders of magnitude more time and money to do this. But it produces order of magnitude better models. How much of the scaling we have now is because of this is opaque but it is significant.
This makes FOOM just as unlikely in 2030 as it did in 2017, because models would still have needed to demonstrate the business case for creating those datasets as they do now. They weren't just "discovered". That limits the FOOM speed.
this is a fair point but I'd counter that even without the AI use case there are business incentives to create and curate well tagged data for marketing and social media uses. I don't want to argue too much in defense of specifically how late the transformer model would need to be discovered to make FOOM possible, the path didn't go as Yud predicted. In theory this there could be some other architecture that would have different features but that's unfalsifiable.
Possibly, I don't really have any insight into the sort of data that is used for these purposes but my gut intuition tells me it is likely significantly more qualitative than the level of quantitative data used for model training. It's likely not QA-pairs + reasoning steps.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link