Here’s another interesting albeit ranty AI post, because not everyone here frequents Hacker News: Punk, or why I don't stream anymore, by George Hotz, who’s a very talented and mildly eccentric programmer.
He addresses the rise of mass commodification and (indirectly here, but a common theme of his other ranty posts) the “permanent underclass”, which has been a problem since consolidation of the internet and economy into a few large companies, but is accelerated by LLMs. (Unless I have bad reading comprehension) “punk” alludes to how the punk culture started novel and authentic, but became standardized and commercialized.
I believe this is a much less discussed concern about ASI, but it’s necessary to prevent a boring conformist and ultimately decaying dystopia. Modern social media and LLMs provide niche subcultures and topics one can explore, but I suspect even these form conformist spirals, because people have become lazier. My idea of alignment would push humans to understand and express themselves (and prevent misery, not sure what else).
Brainwavez
27d ago
The exact outputs usually aren't in the training data. Although similar outputs are, you can take any human idea and decompose it into similar older ideas and maybe an infinitesimal amount of chance. That doesn't mean AI will reach human-level intelligence, but makes it impossible to disprove.
We already live in a human zoo ruled by Moloch.
Why do big companies and governments follow the same common patterns? If you're the leader of a successful company and don't follow those patterns. you'll be pressured, if that doesn't work fired, if that doesn't work your company will go bankrupt. Same with politicians and governments.
Likewise, why is social media so mediocre and toxic? I don't actually think it benefits the rich (I can't imagine how) but is a natural consequence of human bias towards negativity. Hence even smaller sites like this are, albeit to a lesser extent.
We are ruled by self-reinforcing cycles comprised of many smart humans. If a participant stops perpetuating a cycle somebody else will step in. These cycles can and will be broken eventually, but either not obviously or not now, otherwise they already would be.
So to me, AI is just another overlord. Although that doesn't mean it can't be worse, the problem isn't that it rules over humans, the problem is how. For example, maybe future AI will be like a smartphone: a virtual assistant that future generations can't function without, that influences their thoughts however it wants based on its advice like a milder Whispering Earing (although current LLMs are unfortunately influencing us towards sameness and model collapse; I hope that if this is our future, the whispering AI still preserves individual uniqueness and anything else fundamental to "good life").
Brainwavez
27d ago
I agree, whenever rationalist types discuss AI, it sounds like they're living in an alternate reality.
Yet their discussions are still interesting and, via insight, occasionally useful. Scott Alexander started (what eventually led to) this forum; him, Eliezer Yudkowsky, and others invented lots of terminology and concepts we take for granted. I doubt they would've if not for the same personality traits that cause them to keep being wrong about AI (mainly, logic over empiricism). A person can't predict anything without occasionally being wrong, or have any good ideas without occasional bad ideas.
AI 2040: Plan A
The AI 2027 authors published a follow-up. Scott Alexander also wrote a separate blogpost and although not in the author list contributed.
It's a very speculative and optimistic timeline of AI's future evolution. It presents five ways or "plans" the US government will intervene. Unsurprisingly, the ASI-pilled authors favor strong, global regulation to ensure alignment. Summaries:
-
Plan A (recommended): the US makes an international treaty with China, pauses AI training (not inference, i.e. no new models but we keep using existing ones), enforces full transparency of future research, then when alignment research advances enough carefully resumes
-
Plan S: the US makes an international treaty with China and pauses AI training for as long as possible
-
Plan B: the US regulates AI at home and demands China also regulate, but doesn't negotiate with them, probably leading to a war
-
Plan C: the US regulates AI and ignores China, so they overtake it and reach ASI first
-
Plan D: the US doesn't regulate AI, we get ASI in early 2031 and it probably kills everyone
Personally, I just don't share the optimism of these guys in either direction.
I think politicians will prioritize culture war and the failing economy over AI regulation, and at most pass some executive orders suggesting companies be more careful. But I also doubt we'll have ASI that can solve the abstract problems "take over the world" or even "keep existing world leaders in power" (they're getting old and increasingly unpopular, their parties may remain in power but only if their policies significantly shift).
What I expect from AI:
-
Basically solve legacy code by rewriting entire codebases, applying very niche domain knowledge, and actually finding and handling edge-cases better than humans
-
Greatly speedup research, leading to new discoveries and inventions. Important but background things like food preservation and medicine will improve from AI-assisted discoveries. Major advancements in math and theoretical physics
-
Much better and cheaper education, therapy, initial medical/legal appointments, personal repairs...maybe reducing but not eliminating human jobs, because human experts will offer these services "premium"
-
Won't replace human artists. Some advertisements and infographics will be AI but even some will still be human. At best it will assist them in a way where the human still fully controls the output, e.g. by generating code leading to new and improved software tools to learn, practice, and create art
-
Used by the vast majority as a personal assistant, but doesn't replace human relations
Yes, to be clear I’m not against taxes or regulations.
I think the important part is that richer and more influential people and organizations should have strictly more taxes and compliance overhead (and not in the sense “both rich and poor can’t sleep under bridges”). Today we have situations where working-class citizens pay more taxes than rich locals via offshore tax havens, and small businesses suffer from unnecessarily difficult regulations (not protecting anyone) while big businesses fail audits then suffer less than it would take to pass them.
More generally: it’s important for everyone to contribute and compromise for society, the problem is when people asking/enforcing don’t evenly contribute and compromise themselves. I think growing lack of effort and honor within the working class should and would be fixed by seeing more effort and honor from their leaders.
Once everyone has to suffer evenly from laws regulations etc., they become more permissive, because everyone wants as much freedom as possible for the choices they want. Almost everyone agrees that (for example) murder is wrong and building safety is important, few would accept safety for living in a panopticon.
If she gave consent over text, then he drove there and she said no, it’s still rape. Acceptable recourse for being led on was to (for example) break up with her.
Brainwavez
29d ago
Brainwavez
30d ago
I think freedom is to have taxation on level of 1 or 2% of GDP like in 1775 USA. I see it as ability to sell and purchase goods and services without constant oversight of taxman. I see it as being able to have contract without tyrannical book of regulations that are impossible to read withing my lifetime. In a sense of freedom, I think that the modern people are completely cucked and lulled into absolute egregious compliance.
Hence my last paragraph: I think the real issue isn’t that freedom (and democracy) are failing us, but that we’re losing freedom and democracy without realizing. Those who advocate for authoritarianism really want more freedom and power for themselves, which they’re more likely to end up with in a functional democracy (for example, to pass and enforce laws for decent neighborhoods).
Brainwavez
30d ago
But is the problem here democracy?
I’m sure most people want quiet, clean, safe neighborhoods. Even hardcore leftists become tough on public obnoxiousness and crime when it directly affects them. It’s possible to pass laws and fund police to enforce them. Some neighborhoods already do.
Why haven’t most places done this? Honestly, it confuses me. Maybe the incentives that select people to run for office and win select against those that target obvious issues in favor of culture war ones. Still, I haven’t seen any evidence that the issue is democracy or would go away in another form of government. It may be enough to figure out whatever the (democratic) places that manage to be quiet, clean, and safe do.
market forces would be constantly fighting against it
Would they? The government controls the market. They don’t have to completely disenfranchise the opposition to maintain a supermajority: just tip the scales so for example the bottom 25% in income (unexceptional) supporter is much richer than the top 25% (unexceptional) opposer.
Adding +1 vote for each child under 18
I think this is a great idea even today, because governments aren’t prioritizing young people enough, it appeals to both sides (conservatives = more children, liberals = empowering women), and has a decent-sounding justification (the baby deserves representation, and the mother usually cares about them, so she’ll vote in their interest).
But I doubt it would offset the imbalance from giving richer people more votes: most people prefer income over votes since it benefits them more directly, so those having more children would be the rich people (with enough income to raise them in living conditions they consider decent, like today).
Brainwavez
30d ago
I think this system makes a underclass strike or revolution more likely, but maybe not, looking at the current system where everyone’s vote counts only in a weak official sense.
The bigger problem with it is that the ruling party can entrench themselves by making it easier for supporters and harder for opposers to earn money. How would you prevent that?
Brainwavez
30d ago
The very fact of having a backup means you have less at stake.
In multiple generations they won’t have a backup, few countries permit great-grandchildren to remigrate.
But even immigrants themselves, if their origin country hates them or is a true shithole, and they can prove this (e.g. they’re an activist), they have no backup, and credible reason to have negative allegiance towards it.
Brainwavez
30d ago
Since they plan to end in 2028, they must be producing some physical disks today.
Brainwavez
30d ago
I also do that: “Russia should be allowed to defend itself from Ukraine…wait, I mean the other way, Ukraine from Russia”
I desire freedom over democracy and realize they’re orthogonal, but in practice, I see no better system to guarantee everyone’s freedom.
How would you ensure the party in charge gives you clean streets and similar neighbors without accountability at least to yourself (e.g. by voting yourself)?
And of course everyone wants to be more free than others, so how would you ensure you’re on top? Personally, I’d do the usual: propose a system based on Rawls’ Veil of Ignorance which looks good enough to convince others but ensures I’m content. Most people are only convinced to join a system if they have some political influence, so to get enough support to put your system into power, usually it needs at least the illusion of democracy.
freedom was considered to be attained when you were under the rule of the (relatively autonomous) lordship of one of your countrymen or clan
Case in point, my understanding is that most feudal lords abused their serfs. I’m sure it was better in small clans and pre-US American villages, but only because everyone knew each other, and more importantly an ordinary citizen could physically confront their leader(s), so they were still accountable.
Brainwavez
30d ago
Then the problem becomes, who gets to vote and how do you decide?
Non-citizen immigrants already shouldn’t be voting. Children of immigrants, maybe if they have ties to their origin country, but otherwise they have no less of a stake in their host country than anyone else.
The median voter will vote for themselves to have other people’s money, unless the selection process is very careful and has a way to prevent corruption (which seems unlikely because the selfless voters will be taken advantage of by definition; it must prevent their selflessness from being misallocated to parasites). Even when voters benefit someone else, they’re still benefiting themselves, just next prioritizing who they prefer over who they don’t: e.g. when high-ranking Democrats support immigrants over ordinary citizens, they still put themselves first. I think it’s more important to figure out why the under-prioritized group (ordinary citizens)’s votes aren’t working to keep their money and fix that; if you only deprive a different group of votes, they may still be under-prioritized (e.g. Democrats may redirect money from immigrants to citizens they think will vote Democrat, but they’ll give no more to citizens they think will vote Republican).
Also, how do you divide the layers? In the US gerrymandering (and FPTP) have locked Congress from doing anything and entrenched ideologically extreme candidates. But I think this easier and something simple would be good enough like: (pick n) 1 candidate for n-2n people and each candidate has vote power proportional to how many they represent.
Brainwavez
1mo ago
It’s not only Roblox but mainly Spyder Games. However I agree: I’ll accept if there’s a different way to fund them that digital games shouldn’t have copyright either, and I already believe Spyder Games shouldn’t be allowed to DMCA Steal a Brainrot copycats.
If one of them is the original individual who used AI to make this character's image, then I'm actually on their side.
Mementum is partnered with the individual who created him, but they’re a corporation based in France. Judging by that and their name, they buy others’ fads to make money, for example by suing others for using them without a license.
There’s a valid argument that Tung Tung Tung Sahur shouldn’t be in any game without permission, and maybe this is less popular, but I still disagree. Steal a Brainrot doesn’t just include Mr. Sahur or make him the focus, he’s one of many characters. I’d rather see more remixes and compilation games like Super Smash Bros without insane corporate backing and licenses. If Tung Tung Tung Sahur’s creator was broke I’d want him to get some revenue from somewhere, but I believe he’s making more than enough from the character’s likeness already, why does he need more?
Actually, I’ll accept if Steal a Brainrot can’t make money off Mr. Sahur’s likeness, by for example selling him specifically. But at least any free media should be allowed to include him without being DMCAd.
And this is why I don’t speak much publicly. “I’m sorry that happened, your feelings are completely unjustified…wait wait I mean justified, sorry again, sometimes I get mixed up you know”.
I’m curious how much other people think (or write or say) the opposite word without realizing. For me it’s frequent: I recall several earlier cases, usually when speaking. I think it’s interesting, a glitch in the opaque process that is human thinking.
Mementum (partnered with Tung Tung Tung Sahur’s creator) is threatening to sue Spyder Games (owner of Steal a Brainrot). Spyder Games is counter-suing Mementum to get official permission to use him, and the screenshot is from that. Spyder Games is simultaneously filing takedowns of Steal a Brainrot copycats.
Brainwavez
1mo ago
I actually forgot the last part: “others that have been tried”.
Maybe in the future we’ll have a better government run by and/or with the assistance of AI. But not today: LLMs are too suggestible, so in practice I’m confident it would be hijacked by humans.
Brainwavez
1mo ago
I agree. Unlike most idealistic proposals (like ENIAC keyboards), I assume most people know what "expansion pack" means, so we can use it in real conversation.
Small "DLC" like cosmetics aren't expansion packs, but we have a word for those too, with a more accurate connotation: "microtransactions".
Even if you buy a PS5 game in a DVD case from a brick-and-mortar shop, it just contains a slip of paper with a download key.
Not always: it was recently a big news headline that Sony plans to end physical disk production in 2028. Personally, I don't think it matters, because the underlying problem that everyone (rightly) points out does matter, not owning your games, is separate (you can buy a physical disk and lose access to the game, and you can freeze and hack a digital game to keep access; it may be symbolic, but symbolism is vastly overrated).
Not true: a group of human brains, or a human + tools|AI, or humans + tools|AI, are smarter and more complicated.
More options
Context Copy link