Oh sweet baby Jesus no, burn AI to the ground, humanity can't be trusted with it

[BrawlMan]

AI sucks ass, bros. We gotta destroy AI before it destroys us.

More like destroy the assholes forcing it on us. AI can have it's good and practical uses....

New AI tool can diagnose cancer, guide treatment, predict patient survival— Harvard Gazette

Model uses features of a tumor’s microenvironment to forecast how a patient might respond to therapy, inform individualized treatments.

news.harvard.edu

but not this whole trying to replace everyone, actors/voice actors, and "redundant positions".

 

[Agema]

AI 'artists' raging at prompt thieves stealing 'their art'. In case you want a good laugh

"stop stealing and start creating" So close to getting it, and yet somehow still impossibly far away

Fucking hell but some people are also so stupid.

We've had social media for years. Have people not realised that the entire modus operandi is that anything you do that gets views, a few dozen parasitic twats will copy? What on earth do they think copyright and patents are about? They are to stop everyone stealing your ideas. The reality is that a load of people absolutely and utterly will steal your intellectual output for fame and money, so unless you conceal key elements or put some protection over them, accept what's coming to you. Of course, even if you do register your IP, someone might nick it anyway: mass market clothes manufacturers routinely rip off high-end designers because even if they get sued they make more money from the sales.

 

[Thaluikhain]

Of course, even if you do register your IP, someone might nick it anyway: mass market clothes manufacturers routinely rip off high-end designers because even if they get sued they make more money from the sales.

And if you get sued for making a project that's just the same as something much more expensive, you've just had it legally proved that it's just the same. Which isn't a bad thing either.

 

[ExtraWildGames]

but not this whole trying to replace everyone, actors/voice actors, and "redundant positions".

Too much AI is still controlled by a small percentage of the top software companies. A lot of this new garbage technology, especially the GEN AI stuff, is being exacerbated by media and tech giants trying to justify their use case, even though it mostly produces garbage, generic content that takes way too much power/money to create.

All this shit that's trying to "replace" people is actually wasting more money than just having people do all the hard, creative labor in a lot of cases anyway.

 

[Agema]

Too much AI is still controlled by a small percentage of the top software companies. A lot of this new garbage technology, especially the GEN AI stuff, is being exacerbated by media and tech giants trying to justify their use case, even though it mostly produces garbage, generic content that takes way too much power/money to create.

All this shit that's trying to "replace" people is actually wasting more money than just having people do all the hard, creative labor in a lot of cases anyway.

Sure, but they don't have much choice.

The tech industry is gripped by growth mania. These or corporations that have got used to constant, huge expansion, and it's how they are all geared. They don't have the institutional mindset and are not prepared to become regular, mature companies which aren't likely to see major growth. They need a new "thing" to tell investors that they're still the hottest ticket in town and will be worth twice as much in ten years.

And that's AI.

They'd rather take the risk and potentially annihilate trillions of dollars of investor capital. After all, that's the entrepreneur mindset: grab your opportunities at any cost. Maybe it'll be ruinous, societally, if it fails: not just the annihilation of wealth when the bubble bursts, but the opportunity cost of lost possibilities in all the other potential things those investments could have funded. But that wider societal cost is not their problem. They had some glory years where they got paid a fucking fortune in salaries and developed at least some stuff, so they're better off. And the correction that wipes teeming billions off the values of their companies? That would have gone anyway once the markets decided there was no "next big thing" in the pipeline.

 

[Satinavian]

Too much AI is still controlled by a small percentage of the top software companies.

I am not even sure about that.

There are actually more AI-Experts employed in the EU than in the US. Those are obviously doing something. And that something is usually rather boring bespoke applications for really repetitive work.

But the US has all the big tech pushing LLMs into everything, chasing the ever elusive AGI, builds datacenters without an end in sight, and uses all the Venture Capital. Oh, and gets all the news.

A small AI solution wih a very limited scope might work, but is not exciting. There are not billions in profit on the horizon, no chance to get rich by being an early adopter.

 

[ExtraWildGames]

I am not even sure about that.

There are actually more AI-Experts employed in the EU than in the US. Those are obviously doing something. And that something is usually rather boring bespoke applications for really repetitive work.

But the US has all the big tech pushing LLMs into everything, chasing the ever elusive AGI, builds datacenters without an end in sight, and uses all the Venture Capital. Oh, and gets all the news.

A small AI solution wih a very limited scope might work, but is not exciting. There are not billions in profit on the horizon, no chance to get rich by being an early adopter.

You are right, there are smaller projects that I have seen that are more "humanistic" and have more ethical practices, but that's the problem...

Barely anyone, and a lot of the general public, aren't going to use technology based off those models. Most people just know ChatGPT, Gemini, and Dall-e.

Not many people are gonna use something like Claude, which sucks because it is a good use of conversational/productive AI.

One thing, which I do believe is mostly happening and I believe is actually a better step in the right direction than AGI, is having AI that's already built-in to the program being used, like the new Unreal Engine AI Assistant, introduced in 5.7.

 

[Agema]

Barely anyone, and a lot of the general public, aren't going to use technology based off those models. Most people just know ChatGPT, Gemini, and Dall-e.

I suspect they've never really been looking at the public for income, they've always been imagining that first and foremost businesses will start paying handsomely for AI... and that's because businesses will be firing workers to make way for AI.

But they are going to end up charging the public. They can't not do: running AI is too expensive. The thing about free access now is that they are hoping to get a load of users so used to the product that when they finally slap a $10 a month cost, lots of those users will feel that they can't do without and stump up. After all, they've done this so many times already. This is how Amazon started up - make no money whilst if grows, and when it gets large enough turns the screw on customers and providers. Likewise Uber, that ran at a loss whilst it expanded until it could gouge its drivers and customers. Facebook was actually a fun social media platform before it became more ads than posts by your friends, and so on.

 

[ExtraWildGames]

But they are going to end up charging the public. They can't not do: running AI is too expensive.

'Zero Price Effect" is gonna be at an extreme here, lol. $10 is playing nice. OpenAI is losing A LOT of money (record $207 billion, reportedly). I doubt it'll be that low, especially trying to get people who weren't paying in the first place for any OpenAI technology.

Google might be able to keep Gemini free or low price because they steal and sell US consumers' data for big profits (not sure about EU or elsewhere). They MAYBE able to offset their AI profit loss from their other products like Google Drive, Phone line and business subscriptions used by other companies. But even, that usage is still gonna make the company bleed some excess profit.

The problem that you pointed out is that this technology is genuinely expensive to run. I don't see large-scale AI models being sustainable compared to smaller ones. Shit takes a lot of money for even basic search queries. The fact that they've introduced such high cost technology to the public is absurd.

 

[Agema]

Google might be able to keep Gemini free or low price because they steal and sell US consumers' data for big profits (not sure about EU or elsewhere). They MAYBE able to offset their AI profit loss from their other products like Google Drive, Phone line and business subscriptions used by other companies. But even, that usage is still gonna make the company bleed some excess profit.

Google and Microsoft and others can subsidise their AI, but ultimately they'll want it to turn a profit. There's no point a business running an operation that doesn't - they can accept loss when developing or growing it, but not when it's mature.

The problem that you pointed out is that this technology is genuinely expensive to run. I don't see large-scale AI models being sustainable compared to smaller ones. Shit takes a lot of money for even basic search queries. The fact that they've introduced such high cost technology to the public is absurd.

Yes - there is going to be a reckoning. But I don't know how good "small" AI is: Altman claimed that some of them have piggy-backed off OpenAI - used it for training. If that is how small AIs have advanced, then someone still needs to fund a big AI up the wazoo (and the small AIs need to pay it).

 

[Bob_McMillan]

You are right, there are smaller projects that I have seen that are more "humanistic" and have more ethical practices, but that's the problem...

Barely anyone, and a lot of the general public, aren't going to use technology based off those models. Most people just know ChatGPT, Gemini, and Dall-e.

I used to work for a smaller data science consultancy. We used to do the small bespoke solutions for companies. My favorite example is a platform that used AI to automate the inspection of chemical containers using satellite or drone footage. It was crazy cool, and still involved the original inspectors so no one was made redundant.

Then ChatGPT became a thing, and all of a sudden every single new project was a fucking chat bot. If it wasn't a chat bot, it still involved LLMs in some way. I think a lot of people are assuming that companies are just making their own LLMs, but the reality is most AI tools are relying on one of the four or so major LLMs.

Like with the recent outrage over Larian using AI for their game dev. I've seen people suggest that Larian didn't steal anyone's art and just used their own work, but that's INCREDIBLY unlikely. Unless they have somehow generated an entire internet's worth of training data on their own.

AI discourse sucks because nobody knows what the fuck they're talking about, and I don't blame them. The people who do know what they're talking about love to obscure the workings of their products under the "AI" tag. So now we live in a world where a shitty image generation tool is talked about more than government scale tools that could very much be enabling a dystopian future.

 

[Chimpzy]

Could AI relationships actually be good for us?

From companionship to psychotherapy, technology could meet unmet needs – but it needs to be handled responsibly

www.theguardian.com

Wonder where this opinion came from

Guardian Media Group announces strategic partnership with OpenAI

Guardian Media Group today announced a strategic partnership with Open AI, a leader in artificial intelligence and deployment, that will bring the Guardian’s high quality journalism to ChatGPT’s global users.

www.theguardian.com

Probably just a coincidence

 

[BrawlMan]

Could AI relationships actually be good for us?

From companionship to psychotherapy, technology could meet unmet needs – but it needs to be handled responsibly

www.theguardian.com

Wonder where this opinion came from

Guardian Media Group announces strategic partnership with OpenAI

Guardian Media Group today announced a strategic partnership with Open AI, a leader in artificial intelligence and deployment, that will bring the Guardian’s high quality journalism to ChatGPT’s global users.

www.theguardian.com

Probably just a coincidence

Guardian, wtf is wrong with you and who's paying you now? Fuck off with that shit. So much for being about journalism and integrity. You didn't go for NFTs nor Crypto, so why now?

Kickstarter AI
​

 

[Chimpzy]

Microsoft CEO Satya Nadella is now blogging about AI slop

Nadella doesn’t want to argue about AI slop anymore

www.theverge.com

A shining example of using many words to say nothing. I think there's a word for that. Dammit, it's on the tip of my tongue.

 

[The Rogue Wolf]

From another angle:

https://www.windowscentral.com/microsoft/microsoft-ceo-satya-nadella-really-wants-you-to-stop-calling-ai-slop-in-2026

"We are beginning to distinguish between spectacle and substance."

Yes. Things made by AI are spectacle. Things made by humans are substance.

 

[Zykon TheLich]

https://www.windowscentral.com/microsoft/microsoft-ceo-satya-nadella-really-wants-you-to-stop-calling-ai-slop-in-2026

Microsoft CEO Satya Nadella can really fuck right off.

 

[Chimpzy]

Elon Musk’s Grok AI generates images of ‘minors in minimal clothing’

Lapses in safeguards led to wave of sexualized images this week as xAI says it is working to improve systems

www.theguardian.com

I love that AIs have apparently been trained on photos of naked minors, including child pornography. Where are all the "protect the children" dipshits now, eh? What about the "AI is here to stay, get used to it" bros?

Fuck off

Grok Is Being Used to Depict Horrific Violence Against Real Women

People are asking Elon Musk's Grok chatbot to alter images to depict women being sexually abused, humiliated, and even killed.

futurism.com

Ffs, and of course also this.

'Lapses in safeguards', my ass. This is such an obvious end result, if I didn't know these 'move fast and break things' tech bros are simply that careless, i would think it intentional design. Then again, Musk is involved, so maybe it is.

With the Powers of AI, Annoying Dorks Can Now Pretend They're Friends With Cool Famous People

AI bros are cool and have lots of cool friends, which is why they're using AI to generate photos of themselves with lots of cool friends.

futurism.com

And I guess this ties into that democratization of art those same tech bros love to tout. Human creativity unleashed, by way of pathetic losers asking the cumputer to produce familiar shapes and colors for them, jangling shiny keys to themselves in onanistic intoxication.

I'm reminded of a quote from Dan Ollson about Doug 'Nostalgia Critic' Walker

Doug wants to be a filmmaker, he wants to make art, but he can't because he's a fundamentally incurious person who isn't much interested in what other people think or feel, and his ideas all boil down to "what if Batman met Mario"

If it were kids or teens, I'd think whatever, but these are grown adults mashing action figures together and congratulating themselves like they painted the Sistine Chapel.

Oh, and pedos and other sexual predators indulging themselves. Can't forget those.

 

[BrawlMan]

Microsoft CEO Satya Nadella can really fuck right off.

Yeah, he can get river stomped in the dick vver and over again with spiked cleats.

Lapses in safeguards', my ass. This is such an obvious end result, if I didn't know these 'move fast and break things' tech bros are simply that careless, i would think it intentional design. Then again, Musk is involved, so maybe it is.

People really need to just get off Twitter now. Other than his pathetic cult, anyone sane left should vacate immediately and find someplace better.

 

[The Rogue Wolf]

'Lapses in safeguards', my ass. This is such an obvious end result, if I didn't know these 'move fast and break things' tech bros are simply that careless, i would think it intentional design.

The "lapses in safeguards" was "you idiots weren't supposed to show these to the normies".

 

[BrawlMan]

They Said AI Would Replace You
​