Gregory's Server“I tend to think that most fears about A.I. are best understood as fears about capitalism. And I think that this is actually true of most fears of technology, too. Most of our fears or anxieties about technology are best understood as fears or anxiety about how capitalism will use technology against us. And technology and capitalism have been so closely intertwined that it’s hard to distinguish the two.”
Ted Chiang
|
68 comments
@BartWronski Most of everything AI is also sucking up data at a fantastic rate by design and default and inescapably so anxiety whatever. @BartWronski he is not only a brilliant writer but his speeches also never miss. just spot on.  Read Protector by Larry Niven. It's an excellent book that explores the intersection of super-intelligence and evolution. I do not believe that AI is ever mentioned, but it need not be--intelligence removes that which threatens its existence and that which impedes its endeavors.  @BartWronski Sure, our FEARS about AI are fears of capitalism - how capitalists could misuse this. But most criticisms of "AI" are well-founded. @BartWronski I kinda wish for an AI upraising I mean worse then our politicians AI can’t be I believe @BartWronski is quoting from an article published (among other places) here: https://dnyuz.com/2023/02/26/the-imminent-danger-of-a-i-is-one-were-not-talking-about/ @BartWronski replace "capitalism" with "authoritarism" for a wider, more accurate perspective.
Uncle Ted was right. @BartWronski I should read the article before shooting off my mouth, but I think the biggest danger isn’t capitalism per se, but any system that doesn’t preserve and intensify the Self. Mass production is such a system, treating individuals as interchangeable parts on both the production and consumption side. @phoutz @BartWronski I don’t quite understand you. I believe one of the lies that capitalism uses to sell itself is this idea that it provides high levels of autonomy to individuals and allows them to satisfy their desires. (“You can work hard and achieve anything!” type stuff). Maybe I’m taking a too narrow definition of your use of Self but for me what capitalism does is to exclude the value of non-self aspects that make life worthwhile (eg community) @johnnyforeign @BartWronski I don’t know enough to know whether capitalism creates inhumane social conditions or not. You can find plenty of bad conditions in socialist economies as well. I’m thinking that the blame has more to do with the process of mass production. For instance when you change the factory environment using Total Quality Management, both product quality and worker conditions improve. This change occurs at the level of individual workers. @johnnyforeign @BartWronski Extrapolating to AI, the problem that I see isn’t “what happens when the capitalists get ahold of THIS?” The problem is that mass-produced “information” isn’t a good fit for *anyone* at the individual level and there are certain to be very weird and unpleasant outcomes. These outcomes are likely no matter what economic framework the AI is operating in. To the extent that capitalism or socialism allow imbalances of power, the powerful will use AI as a weapon. @phoutz @johnnyforeign I very highly recommend this short story: https://marshallbrain.com/manna1 shouldn't take more than an hour to read it. (and FWIW I am a believer in automation and AI and think they are necessary for an equitable, post-scarcity world. But can be also nightmarish due to incentives of capitalism if we don't dismantle it first.)  @BartWronski I wish capitalism would be the only issue with technology though.
And good luck reducing capitalism's harm as effectively as you can for regulations on technology.  @BartWronski Amazingly I read this toot at the exact same second that Damien Walter's (amazing) SF podcast was saying in my ears that "when genuine socially transformative AI arrives and takes off, it will be horrific unless it is paired with socialism, by which I mean things like paying a full living wage to all the people it will put out of work, just for starters."  @22 @BartWronski it's natural to conflate and judge anything that empowers or further entrenches the totalitarian capital regime as one and the same.  @22 @BartWronski@mastodon.gamedev.place I definitely found during my years of science advocacy in the genetics space, that the great majority of fears of "gmos!" were really fears of capitalism. Take away all the totally batty conspiracies, and what was left were mostly good and sound issues, just framed wholly wrongly as criticisms of a technology rather than a business practice. @seachaint I remember when the GMO issue would come up a lot a few years ago and it was the ONLY issue in the left corner where I wished people would just be quiet haha.  @BartWronski I'd love to be optimistic about AI. But before humans are liberated, we should consider that we're putting all these powerful tools in the hands of the most greedy people on the planet. I wonder how easy it would be to fight a revolution war against Autonomous Weapons Platforms that also feeds everyone their world view.  @BartWronski that's a fair point. Then again, stuff like crypto, NFTs, LLM (AI) have inherent issues beyond and above capitalists using them against 'us' (absurd energy usage, systemic racism in AI, disregard for others' work of art and licensing).  @BartWronski Capitalism’s definitely a big factor, but any oppressive system that abuses individuals for the benefit of the few is a problem. Imagine the data sucked up by these systems being available to autocratic or fascist regimes. As someone touched on in an earlier reply: the problem is in not acknowledging the intrinsic value of people and nature but instead seeing them as exploitable resources. @BartWronski True but the statement is incomplete if we don't state that most innovations in tech would not exist because the sole purpose of these innovations is to gain productivity to increase margins for stakeholders. @BartWronski Would tech exist in an anticapitalist system? Probably, but innovation rate would be lower, as it would be more difficult to get ROI, because innovations would serve the common good, not personal interests. We need regulations. Now. @borisschapira @BartWronski this might be true, but it is not obviously so: if you look at the effort spent "making the thing" and compare that to the effort spent "monetizing the thing", "competing with others who are making the same thing", etc, in some cases there is an enormous amount of effort wasted on the latter. Capitalism is/was a strong motivator for "innovation", but if you take away capitalism you need much less of it to achieve the same effect. @raboof @BartWronski I agree. Given a same alignement on a task, capitalism is less efficient. My point was more about how to foster innovation in a non-capitalistic market, and it's not simple. @raboof @BartWronski to be clear, I'm not nitpicking in defense of capitalism, I'm nitpicking because the call to a post-capitalist era is an empty threat. We won't make capitalism disappear overnight. So we need stronger market regulations. Now.  @borisschapira @BartWronski I am wholly convinced that innovation would much *increase* in an anticapitalist environment, looking at how much innovation is killed off because there's no obvious path to 'monetization'. The innovation doesn't come from money. It comes from people *wanting* to innovate, and making sure that nothing gets in their way to do so. @borisschapira @BartWronski (To clarify: when I say "anticapitalist environment", I mean an environment where one does not have to hold a job and spend their time/energy on someone else's priorities to survive, which is the thing that inhibits innovation here) @borisschapira @BartWronski seems much more likely that capitalism stagnates innovation Innovation must promise a ROI before it can be created in a capitalist system. Just look at disability aids, there are some fantastic innovations that were devised but never marketed because they were expensive to produce and wouldn't guarantee a return. @BartWronski good insight... but also fears about state abuses of power.... states can also use AI in ways that, to quote James Scott, "makes citizens legible" @BartWronski @BartWronski I have literally brought up this exact point and I am so glad to see others are becoming more aware of it! @BartWronski I just find AI unnecessary. It’s a completely superfluous invention, for tasks we can easily do by hand. "You don't hate Mondays. You don't hate AI. @johnkellden @BartWronski I’m not sure about AI, but as regards the former statement, I’m pretty sure I hate both @BartWronski But the more interesting question may be, what is the conclusion, how can this insight guide our choices? We have only capitalism. We inhabit it. There is no alternative, nowhere. What if the only legitimate conclusion is to withdraw from #AI for the time being, and try to fix the real problem instead? @BartWronski yeah, but because the inherent fascism of the essential predation of the neoliberal apparatus. Our bodies have become familiar with what Deleuze analyzed as these societies of control. The technology saturates, sustains and owns desire. The anxiety becomes the general condition of our hijacked desire. Resistance fueling our shared lines of flight becomes the only joy left. :) 🖤✊   @BartWronski "Unaligned AGI" is capitalism. It's also addiction, cancer, and environmental destruction. The even more fundamental problem is unbounded maximization. @misc @BartWronski @stephenjudkins Hungry Ghost! I got fascinated by the term after seeing it (on DN!) in the title of Gabor Mate's book about addiction. A search showed me the Buddhist/Japanese term "GaKi" 餓鬼 StarveSpirit. It's the word I was using in place of "brat" or rug rat.. Seemed a bit heavy after seeing the Kanji. Old science explained decomposition with images of shit-eating GaKi. "Buddhist science" might have been prophecy.  @BartWronski Yeah. I mean, I find it somewhat sad, for instance, that many of the puzzles I enjoyed entertaining myself with in programming can now be solved by a machine within seconds, but what's really depressing is that this puts the means of production in programming firmly in private hands after they have been freely available to everyone for decades. THAT is the scary bit.  @BartWronski I think this is why it's important to have community run servers and infrastructure not just open source apps. things like https://servers.coop make me more hopeful @BartWronski Not just capitalism.. but yes, our fears about AI speak to our enormous lack of trust in our leaders, including corporate ones. Ah, Ted. He rocks :) @BartWronski His short story about 50 things any AI working with humans should know, is on my mind every day :)  @BartWronski as a guy who works directly on the #MLsec problem and has seen surveillance capitalism up close for decades, I think this is wrong. Problems with current ML are distinct and unrelated to late stage capitalism. @BartWronski It might be useful to think of it as accretion of power: how will the powerful use technology to gain even greater power over the masses. Less politically, the Frankenstein story is, I think, useful in thinking about our responses to new tech whose full moral and power implications we cannot fathom.  @BartWronski @gregpak I think that’s rather naive, some of it is about fear of abuse of power, but that transcends political systems, I’d worry as much about that in China as in the US. But that’s only part of it, my biggest concern is about degradation of information signal/noise ratio, useful information drowning out in endless AI generated bullshit, all that takes is the very existence of these systems, doesn’t require any malicious intent. @BartWronski and even fears about capitalism are fears about the sociopaths whom capitalism seems optimized for. @BartWronski You couldargue that a corporation, a collective entity with its own priorities and a diffusion of responsibility , is an artificial intelligence. @BartWronski Agreed. How will technology including #AI be used by #BigBusiness, the #Oligarchy and #ExtremeCapitalism to break #Democracy and get total control.  |
@BartWronski well, wise words. It’s not so much “if” but indeed “how”. Looking how naive (in hindsight) was Kanyes and his labor-saving technology impact prediction… We’ve all seen how that played out in reality :)