Why are the AI Companies spreading F.U.D. about AI?
Posted by supracode@reddit | LocalLLaMA | View on Reddit | 20 comments
A couple of recent videos I have watched :
Billionaires Are Funding 'Anti AI' Content
AI Manufactured Doubt (long but interesting take)
My tin foil hat take : AI Companies understand that offline llm hosting is becoming more viable for both individuals and companies. They are spreading the "AI is dangerous" message to get government regulators to pass laws to keep the people "safe" from the unbridled power of tokens and weights. They will use their lobbying with the FUD as ammunition to pass the "AI Safety for the Children Act" to keep their grip on a soon to be commoditized industry.
Am I crazy? Maybe I have AI Psychosis?
mantafloppy@reddit
Ai CEO made big promise for massive amount of cash, in exchange of saying they would develop AGI.
Now that they cant deliver, they wanna be able to say its not their fault to the investor, its the gov that dont let them do it for "security" reason, and keep the money.
mohelgamal@reddit
The truth is very simple and not a conspiracy at all.
The real goal is just to IPO for an ungodly amount of money, their main to sell AI to corporate leaders who want to cut labor costs. They are telling the CEOs “AI is so good it is gonna make your productivity explode”
Corps don’t want to spend money to merely make their employees life a little easier, they want to be able to replace people because people are expensive
So now AI CEOs have to hype the shit out of the idea that AI has god like powers, world changing powers. It is gonna cause mass lay offs and and leave the world with a few oligarchs and hordes of starving peasants. Do you want to be an oligarch or a peasant ?
They don’t care that 90% of people fear an “apocalypse” they know is not likely to come. They care that every person with money would want their stocks at the IPO. After that it doesn’t matter if AI takes a year or a century to change the world. Their lives have changed
When Anthropic had their tiff with the pentagon, it was widely profitably for them because, sure they lost. Government contract, but delivered a message to every CEO in the world that Claude is so freaking good, it is running US defense so it sure will help you run your warehouse better.
TBF, AI does have some concerning power, I personally feel a little worried that someone may use an unrestricted LLM to make nerve gas, or guide some other form of mass murder in their basement. It is kinda of like gun laws, most gun owners are law abiding, but when every twitching lunatic can walk into a store a buy an assault rifle is not really a good situation.
JamesEvoAI@reddit
Money, the answer is always money. The question is never why, it's how many steps do you have to trace through before you figure out who is making the money.
xienze@reddit
My tinfoil hat take, when you see videos that come out of nowhere and rapidly gain traction proclaiming that "billionaires", "China", "Russia", or
So in this case, I think videos like these going "whoa did you know it's really billionaires spreading fear about AI and regular people haven't organically come to the conclusion that AI might be bad for their livelihoods, electricity prices, hardware prices, and environment???" are probably being funded Anthropic, OpenAI, and the very billionaires investing outrageous sums of money in AI. The point is to gaslight people into thinking that AI skepticism and anger over the very obvious bad things that AI is responsible is totally irrational and, if you think that, you're no better than some evil billionaire or Russian!
supracode@reddit (OP)
So it's almost like AI FUD Inception... AI Company says AI is dangerous -> AI company pays for FUD Videos that AI Is Dangerous -> AI company pays for videos that say that AI Billionaires are promoting that AI is dangerous? All to make people ignore the real issues like upcoming water and electrical shortages... I like your thinking!
Pleasant-Shallot-707@reddit
oil companies did the same thing with Fracking
Kahvana@reddit
Generally found these type of topics not productive; it invites doom and gloom of things outside our control besides voting. I rather spend the energy on things where we can make a difference, instead of worrying over it.
Download and back up the weights you want to have, and make sure you can run everything locally.
Borkato@reddit
You’re not wrong. I’m very, very prone to anxiety and honestly it’s so true that all of this damn mental energy is better spent doing what makes you happy while ensuring you can still do it through voting and staying informed every now and then.
MrPecunius@reddit
All companies try to erect barriers to entry if possible, so of course they are in favor of regulation.
DinoAmino@reddit
Maybe it's just the medium you're in? Vids are all about hot takes, generating either excitement or outrage or both, and attracting eyeballs. Fear sells.
ambient_temp_xeno@reddit
Youtube went from early slop to actually good during the money bonanza, then utter slop that made me nostalgic for the early slop.
mobileJay77@reddit
I don't know, but that is one reason I like my local stack. I can observe myself.
AI will consume all water? Maybe you need better engineers for your data centers.
Embarrassed-Area4652@reddit
Heard someone call it “critihype” the other day: if the “critique” of something is that it’s too powerful, it’s going to put us all out of a job, it’s emergent and has become artificially sentient without anyone realizing, etc., that’s not a critique.
Tired__Dev@reddit
The use of regulatory capture in tech became huge during Cambridge Analytica and the 2016 presidential elections. The Zuckerberg Congress/senate hearings were that moment where Zuck got to standup in front of technologically incompetent politicians while they ask him the dumbest questions. I’ll never forget after something 3 hours of questioning one of the senators was like “we should take a break” and Zuck said “I can keep going” with a dumb smile on his face. Then there were Netflix documentaries that said absolutely nothing about how these companies are spying on you, but basically made the whole of Reddit conclude that regulation was needed.
L As Zuck showed in his hearings and said on multiple occasions he was happy to get his team with legislators to fix “the problem”.
I’ve seen all of this shit behind the scenes. All of these regulations are theatre and really just a pay to play so monopolies can maintain their market share by building a regulatory moat.
Now we’re in the AI bubble and open source Chinese models are 8 months behind. We’re hitting the wall when it comes to adding parameters and most of the real work going on with these companies is now providing context to the model with traditional software. There is no moat and that’s dangerous to the world economy. So of course they’re creating fear around what they’re building and then telling politicians that only they can help while China makes them look stupid.
The real end to this is the bubble bursts, and if there’s a world economy left then hardware price hit the floor like they did in the dotcom bubble. I suspect then there will be a rationalization of the tech and smaller upstarts coming back. People will provision open models in the cloud and maybe OpenAI and anthropic will just be what Oracle was for a bit if they still exist.
MoneyPowerNexis@reddit
Hello AI company! I am going into production and I want your strongest models!
utxohodler@reddit
My models are too strong for you, traveler.
Demortus@reddit
Because they want to reduce competition from open source models, which would be disproportionately affected by regulations.
GrokiniGPT@reddit
you might be on to something, why would the big companies want open source models available when they could try and obtain a monopoly? good stuff
supracode@reddit (OP)
https://financialpost.com/financial-times/ai-guardrails-meta-google-models-minutes
Maybe they let them out, and say... "We can't allow this to happen!". Somebody might use the LLM to find out something that was already on the internet!
ZenaMeTepe@reddit
The real question is, can LLM companies survive even in a scenario where they completely monopolize cloud inference and make local models illegal at most levels? I don't think they can. They'll try tho.