What do you think about AI, overall?

[digdeeper]

Let's have this be the first topic that I post specifically here. As I just had this thought, and figured might as well. Anyway...how hopeful are you about AI?

As for me, I obviously know about the "job stealing" issue that it's already bringing in, and that the gains are siphoned by the already very rich, and we're totally unprepared. As well as danger of people - who over-rely on it - becoming lazy and stupid. Obviously the issue of artists losing a way to support themselves because the AI companies just grab it all, with no care for copyright or IP law (not that I like this, but it shows that it doesn't apply to the very rich and powerful at all). I know the issues with "slop" overrunning search engines, social media being now for the bots and not the humans, video games having badly made AI art in it to extract the last bits of profits from nostalgia. I realize the creepiness of resurrecting dead people by AI and fooling the family it's the same. The danger of autonomous weapons is obviously high. But does it mean it's all bad?

I mean, obviously it solves certain problems more efficiently than humans. For a simple example you can go look at certain games that it already has outcompeted humans in, like chess. It finds solutions for speedrunning that humans might have never (https://inv.nadeko.net/watch?v=zFLQU70QstY). LLMs are already very good at translations, maybe not yet at the top-human level, but better than the average person that finished learning a foreign language in school (I mean here primary and secondary). They can often give decent health advice and I'm surprised at how well they can work as therapists (no, I don't use them this way, but I know others who do). Of course they can also give disastrous advice - but so do humans. And this makes therapy and advice in general more accessible. And yet, media only picks out the bad outcomes and not the probably more numerous good ones. And this is just LLMs. What about all kinds of other AI, like one that could actually learn and interact with the real world?

Of course, the usual problems with it haven't gone away. But it seems that many of them could simply be fixed with a better economic and social systems. And this is where I come to the application for AI that I am the most hopeful for. Basically, it is clear that human rule has so far failed and generated massive destruction and so on. I mean if you've read my site you will know. And I'm not saying this as a given at all, but perhaps AI would be able to bypass some of the human tendencies (or at least, the tendencies of the people in power) that have been ruining the world for decades or more at this point. Basically the people at the top are extremely cruel, seemingly stupid, and completely disconnected from the outcomes of their ideas. And they have brainwashed most others to believe what they do because of the power they have over the media, etc.

And my question is: could AI fix some of it? If only it was given actual power over physical reality, not just word shuffling. Could it analyze the bullshit laws we follow and find flaws in them and suggest alternatives? Could it find more efficient ways to allocate resources? Could it find solutions for issues in the human world the way it finds them in the video game world, that humans would never? Could it do it without "special interests" trying to fuck with it? Could it analyze our education systems IMPASSIONATELY and find the cruelty in there, without reference to human inertia and attachment to the current schooling system? I think even current AIs would be able to expose a lot of the dumb shit we do. But the true breakthrough would be if it didn't limit itself to just "exposing", but be able to apply the changes it comes up with, towards how stuff actually would work, without being able to be stopped by the psychos in power. Internet people are really scared of a robot getting out of control and doing something really destructive, but humans have already been doing that for so long without any barriers! Maybe it's an unjustified fear.

And I am asking this because I'm getting increasingly disillusioned with the idea of human rule and human ability to fix the bullshit they themselves shat out. Not that I believe AI rule would be perfect or even good either. I am just not very hopeful about humans breaking their own chains at this point.

But now it's your turn. How hopeful are you about AI?

[sam]

I dislike it, mainly because it's been forced into everything. It seems like the things it is currently capable of doing has made those things more accessible but at a cost to those who truly care about such subjects, as they are now surrounded by idiots (programming is one such case that comes to mind). AI articles and stuff are already everywhere and I can barely even tolerate reading them (if I do at all). LLMs just tell you what you want to hear, we have managed to make computers capable of communicating, the problem is that destroys the entire point of communicating, talking to another person. Neutral networks themselves can probably be pretty useful and have been getting used LOOOOONG before LLMs showed up with their huge waves of slop. It also does not help that most of these Ais have very poor training data, it's been trained on the type of people that use it so they don't seem to be very high caliber.

[TheGreatDeal]

I think most people will soon become deeply disillusioned with AI.

Every day, it generates an incredible amount of junk content, displacing genuine creativity.
It's used for videos of Stephen Hawking playing basketball. This consumes insane amounts of power and electricity.
It will obviously take away jobs from many people, since almost all businesses and all politicians are guided by simple solutions.

It won't benefit citizens, since only corporations and governments will be able to fully utilize it. And they always do so only to increase their power, money, and so on. If those who ruin the lives of ordinary people at least sleep and don't do it 24/7, and are even too lazy to ruin people's lives during work hours because they don't want to work, AI will seek the most effective ways to suppress the population 24/7 in the interests of a small group of people, since the only right in the world is the right of the strong. And it will become almost impossible to resist.
Moreover, it will be introduced into the primary education curriculum, making all children idiots, and from there it will snowball. Society will be divided into those who will be able to receive services from real doctors, teachers, etc., and the plebs who will live online.

Naturally, these aren't all the downsides, but only the obvious ones.

[qualia]

I run a local LLM and have found it personally useful for menial, repetitive tasks like writing formal emails and generating boilerplate code that I would have needed to mindlessly type out or copy from somewhere anyway.

I have not experienced many instances of AI being forced everywhere as I (almost) completely avoid proprietary software. I have heard of shit like Microsoft pushing Copilot into the fucking Windows notepad which is so terrible it becomes hilarious.

I do not think it will take many jobs away entirely. Maybe I am too optimistic about this aspect but looking at how its already being used for things like writing news articles or programming and how its terrible even when compared to modern journos and soydevs makes me think that actual competent professionals will keep their jobs while the only people who will get fully replaced are the people who deserved to be replaced anyway.

What I'm worried about is the prospect of people who graduated from ChatGPT University getting into professions on which human life depends. Imagine being operated on by a neurosurgeon who passed all his exams by cheating using ChatGPT glasses or something. My only hope is that most of these people will become unhireable due to their lack of knowledge but looking at how many incompetent morons are there in prestigious jobs already, I somehow doubt it.

When it comes to "displacing human creativity", as one user above put it, I'd argue that if you are using AI to shit out slop content you had no human creativity to begin with. The people who are making this shit would never make anything without AI.

There will always be a place for real art. The "artists" AI will replace are pop "singers" who don't even write their music or lyrics and scriptwriters for "Fast and Furious 23".

But ignoring all this speculation about the possible future the thing I see people most often bring up is the usage of AI for mass social media political propaganda, information censorship (when used as a " search engine" of sorts), scams, spam etc. While this worsened with AI, scambots, propaganda bots, search engine censorship etc. have all existed long before AI. The only difference is that back in the day guys from Delhi used to run the scams, while guys from Langley ran the propaganda. These days the guys from Delhi do all of it for some measly (measly from 1st world perspective that is, its big for them) financial gain.

The mainstream internet has been shit for a long while and would've been shit with or without AI.

[sirfessor]

So diggy, you're pro technocracy, isn't that what they want and the tech bros? Machine learning is programmed by humans so it will do what is programmed by them. The ones doing that right now are the big tech companies.

The only thing that concerns me is it's going to make surveillance more easier.

I myself pretty much don't use anything A.i, except just recently i found a website where you could put your sentence and writes it in a natural, fluent way. It's definitely great since i can put my words there and it produces something way more coherent than I could ever imagine. Still, never use it on online since i feel its kinda cheating and it could make my writing even worse because i would be relying on it instead of my own brain.

[equilibria]

.

[Admin]

We need more technology floating around allowing idiots to be confirmed in open spaces. When I see someone speaking into a box that spits out biased people pleasing information, I know I can write that person off completely.

[digdeeper]

So diggy, you're pro technocracy, isn't that what they want and the tech bros? Machine learning is programmed by humans so it will do what is programmed by them. The ones doing that right now are the big tech companies.

No, quite the opposite. I know the people in power won't give up the power. But I think the AI might turn against them at some point.

Machine learning is programmed by humans, sure, but it doesn't always do what humans desire or can even predict. Just like the chess engine won't always make the move you think it will.

Basically the people in power will always do evil, but the AI might do something good, if it is given the opportunity to.

This thread isn't my shilling for the AI, but putting it up as a possibility for change, because I doubt humans have what it takes, at this point.

[GenericKeyboard]

The job market, the music and entertainment industry, social media, etc, have been dead far before AI as we know it arrived.

[equilibria]

.

[equilibria]

.

[Theundercoverman_]

The only thing AI is good for is CPU opponents in video games. Everything else I wish never happened, and the worst thing about it is, as others have said, it's being forced onto everything and people will eventually become so reliant on it they can't imagine living without it, like they have with smartphones.

[www]

Okay, I think artificial intelligence killed knowledge gained among people on the internet. Of course there is some good side but if artificial intelligence does not deliberately cause any problem for big companies and I hardly think is going to happen, I believe world will be more dark and guarded. Remember AI won't be just within WWW....

[travolter]

Since the brainless zombies are using they smartphones to scroll their screens forever..... there is no time anymore to study, learn, or spend time thinking... and create new original things.

AI is not here to destroy society..... society was already destroyed since last 10-15 years

[moeloli]

I could write pages worth of barely coherent ramblings, but to keep it short/organized:
- For the regular user, 99% of AI's "use cases" are just vapid novelty entertainment or stuff that you could do by yourself - AI does it more "efficiently" at the cost of lower quality and making you dependent on it/atrophying your brain.
- For the government/corporations, AI is/will be used to push dystopian bullshit in various ways and to manipulate truth.
- If you unironically (or "ironically", whatever that would look like) use AI you're spiritually a street shitter pajeet akin to "people" obsessed with goyphones, etc.
- Local/open source AI is only "better" than capitalist centralized AI as much as degoogled goyphones are "better" than stock goyphones.
- The fact that AI "taking away jobs" is such a public concern brings to light how fake and gay the capitalist system and jobs are.

Anyway, the reason why I wanted to post in this thread is that I've heard news that Sora AI (OpenAI's video generation shit) is shutting down for customers.
Many people would see these news and probably think "nice, we'll have less AI slop and disinformation!", but in reality, that just means that (((they))) are taking away the weapon from the goyim and keeping the fake news machine to themselves.

[sirfessor]

After trying duck.ai, i think translation jobs is definitely going to take a hit. AI translates really good with just a couple of mistakes that i had to fix and what's more the translated langauge was in dialect and not used a lot on the Internet, unlike English. I wonder how it got trained.

[qualia]

After trying duck.ai, i think translation jobs is definitely going to take a hit. AI translates really good with just a couple of mistakes that i had to fix and what's more the translated langauge was in dialect and not used a lot on the Internet, unlike English. I wonder how it got trained.

Ai at the moment is good for translation but it wouldn't be able to translate something like a novel as well as a human translator, and I doubt most serious organisations would be willing to trust it with legal documents and the like. this might be merely a matter of social distrust though and in a couple of years this might change, i can't claim any legal knowledge to judge the validity of ai translations

also, if you have a decent enough gpu check out running your local llm with https://ollama.com/ . id be distrustful of duckduckgo for reasons im sure youre familiar with

[moeloli]

By happenstance I found that Ecosia (the search engine that claims to help the environmentarino and plant trees with the money it makes from ads in search results), which I never took seriously (or cared/spent mental energy thinking about, really), joined the AI bandwagon like everyone else and now has a robokike chatbot, and sometimes it shows automatic robokike summaries on search results like other search engines, despite AI being infamous for its energy consumption and being bad for the environmentarino. That really exposes the search engine as just a joke, something for hipsters to LARP with.

[LoadingXML]

Warning: what you will read is written by a west hater, have some understanding before you get angry, thanks in advance.


I think more of long term benefit A.I is giving us indirectly, let me explain:

After A.I came, many social and economy changes happened that seem to lean towards my dreams/goals, such as:

accelerating the destruction of the west; by making western WI dumber than they already are, A.I is proven to smooth your brain and destroy brain connections, thus the longer A.I exist, the more the west fall.

Sadly i dont think this effects europe as much, lets face it, europe is like in stone age compared to U.S, the fact there is no european A.I or A.I company that challenges the big players is a roof of this.

The second best thing is china just ruining the dinner for the west, as A.I companies burn stupid amount of money to make next big A.I, china (deepseek) just drops down a model for free, thus completely ruining the monopoly the west seems to aim for.

If this trend continues, basically the west would have spent billions to gain nothing, as people simply could run AI locally with no cost.

But even, assuming its not the case, and A.I companies did make westrrners buy A.I access, it will just do the smooth brain effect.

Basically a win win situation for me.

The only bad/good effect I see A.I did was killing the entertainment crap.

Artists of all sorts found themselves jobless, as generative A.I could do their jobs in seconds, this is good (in sense evil people, the artists, just got punished for their sins) but also bad because these artists wont remain jobless and possibly go and do an actual useful thing for the economy.

One could argue this effect is balanced by the fact that more workers = cheap labor, but i doubt number of artists surpass number of women (that suppose to be in the kitchen) which had the biggest effect of labor cost (go figure, reason you are paid less is thanks to feminism).

[moeloli]

Google is going to kill its search engine and fully replace it with the robokike. Welcome to the gay new world.

[qualia]

While Google says that AI Mode is not the default experience, Search’s user interface encourages users to ask follow-up questions instead of scrolling down to the links to other pages.

Normal search will apparently still be a thing (wonder for how long though?). I bet there's going to be a lot of backlash, even from normies, hope it becomes one of the 90% of jewgle products that get shut down

[qualia]

https://inv.nadeko.net/embed/4WyduoGpIPo

Seems normie youtube channels are already picking up on this, this is a massive channel with 15M subscribers talking about how you need to stop using google because of their ai "features"

[LoadingXML]

how you need to stop using google because of their ai "features"

One cringe thing that happened to me recently, I use Google search (surprised?), and using Ublock origin I removed the stupid A.I search summory, and was like done with it.

Google search always had that list like panel in which it tries to highlight some site details relevent to your search, in many cases it is useful and all.

Recently when I open one of these lists (I think only top one), I get "A.I overview".

I don't know if it is right to call this disrespectful really, after I blocked the A.I blob at top, it came to hunt me at the bottom.

Anyhow Its about time I won't press the once useful highlight list...


Bonus: ok ok you wanna know why I use Google, basically despite all its stupid search results it is still better than daffyduckGOY, or any privacy "friendly" search engine, for me if I had money and was in better situation I would have paid for kagi, but muh.

[digdeeper]

Anthropic CEO wrote something very concerning:

Thus, for downstream applications of AI—in contrast to AI itself—I am more worried about the regulatory apparatus slowing down progress (because it can’t handle the increased pace of change) than I am about it failing to address important risks. The last thing we want is for the benefits of AI to be slowed while its risks loom large, so it’s important to take action on this problem as soon as possible.

Really??? We're on the verge of a world where everything we do is tracked, saved into a giant worldwide database, connected with all the other things we've done in our lives, analyzed by a supercomputer cluster to predict what we're going to do next and possibly apply punishments, and "slowing down progress" is what you are thinking about?

Beyond these specific examples, agencies should also consider more radical and flexible mechanisms for accelerated approval. If my predictions about AI are correct, there will soon be many instances of interventions that work really well out of the blue, and the regulatory system should be prepared to take them seriously and not adopt a posture of excessive skepticism.

"Accelerated approval" worked great for the Covid vax, truly.

In theory the article is about mitigating AI dangers, which he does understand (some of them anyway), but his "solutions" are a joke. For example:

Create reliable accountability rules for fully autonomous weapons. Autonomous weapons, and especially any autonomous systems that coordinate or direct them, should be required to respond to mechanisms of constitutional and command accountability (e.g. court orders, legislation, and accountability to senior human overseers) rather than blindly following orders. This could mean that a suitably-designed legal review panel or the judicial branch have their finger on an “off switch”, that the systems themselves are intrinsically trained to seek out and respond to legitimate oversight authority, or both.

How well do the "accountability rules" work today? Innocent people end up in prisons constantly, or even are executed, etc. How well is Zelensky accountable for locking up millions of males inside a war zone?

Ban the domestic use of fully autonomous weapons. While there is a legitimate case for the necessity of fully autonomous weapons to defend against foreign adversaries (such as Russia invading Ukraine), there is no justification for their use against Americans. The military already has some limits on its ability to operate domestically, but ideally these weapons should be banned in law enforcement as well.

Haha...and he's a Zelensky worshipper too. And an American exceptionalist.

In addition, if powerful AI enables deeper and potentially permanent forms of autocratic repression (see Section 4), this makes it all the more important that the world’s most powerful nations are democracies—or at least that strong protections exist against AI-driven repression. It also increases the urgency of a focused geopolitical strategy.

We are already "democracies" which is exactly why plebeians have no effects on what happens. What is needed is direct action against data center building until we can put the decision making about this tech to the people most affected, or maybe eliminate it altogether since it seems humanity isn't ready for it.

Democracies should seek to form a global coalition centered on building AI according to their common values, iteratively trying to draw in the rest of the world by making it more and more attractive to be part of the coalition and less and less attractive to be outside it. The coalition should be a coordinated internationalization of the AI policy ideas discussed in Section 1 through 4, plus an effort to lock down the supply chain critical to building AI by sharing it within the coalition and denying it to those outside it. Some principles and operating goals might include:

Yeah because "global coalitions" such as the EU work so well currently..."Common EU values" sacrificed hundreds of thousands Ukrainian males.

Mutual defense. Countries in the coalition should work together to defend each other with AI and from adversaries’ AI. The coalition should collectively ensure sufficient production of AI-led cyberdefenses, AI-powered drones, AI-driven manufacturing, classified AI compute, AI-driven R&D, and sharing of AI-driven intelligence collection.

He's still thinking in terms of some "outside adversaries" that would also use AI, to justify AI warmongering. When the actual solution is to dissolve states and multinational big corpos supplying AI weapons. I mean let's think about how insane this type of thinking is. You worry that some other big state will develop AI weapons to wipe you out so you want to develop better weapons, instead of destroying the entire system that allows this game.

I'm of course cherry picking the quotes. The entire article isn't that bad - at least the author recognizes some of the dangers. But clearly this guy is at the very least very naive if he believes modern "democracies" will use AI for anything other than plebeian abuse (like every other tech). As long as AI fits into the already existing systems, then it's just another enemy tool to fight against.