this post was submitted on 16 Aug 2023
180 points (100.0% liked)

Technology

1085 readers
9 users here now

This is the official technology community of Lemmy.ml for all news related to creation and use of technology, and to facilitate civil, meaningful discussion around it.


Ask in DM before posting product reviews or ads. All such posts otherwise are subject to removal.


Rules:

1: All Lemmy rules apply

2: Do not post low effort posts

3: NEVER post naziped*gore stuff

4: Always post article URLs or their archived version URLs as sources, NOT screenshots. Help the blind users.

5: personal rants of Big Tech CEOs like Elon Musk are unwelcome (does not include posts about their companies affecting wide range of people)

6: no advertisement posts unless verified as legitimate and non-exploitative/non-consumerist

7: crypto related posts, unless essential, are disallowed

founded 5 years ago
MODERATORS
 

Investors are barely breaking even as the venture is hardly making any profits due to a shortage of chips, divided interests, and more.

... OpenAI has already seen a $540 million loss since debuting ChatGPT.

... OpenAI uses approximately 700,000 dollars to run the tool daily.


⚠️ First off, apologies as I didn't cross check. Take it w/ a grain of salt.


This piece of news, if true, somehow explains why OpenAI has been coming up w/ weird schemes for making $$$ like entering the content moderation space.

On a similar note, I wonder if this had been a key driver (behind the scenes) in the recent investment in open source AI initiatives (Haidra comes to my mind?) Perhaps some corporations who haven't got enough $$$ to fund their own dedicated research group are looking to benefit from an open source model?

top 50 comments
sorted by: hot top controversial new old
[–] Clymene@lemmy.ml 55 points 1 year ago (2 children)

Too much is made of the shrinking user base. I’m sure they’ll come back with a vengeance come the start of the school year in the northern hemisphere.

Also, maybe a tool like this shouldn’t be privately funded? Most of the technology is based on university funded research we all paid for. mRNA vaccine research was similarly funded with public money in mostly universities, and now we have to pay some private company to sell it back to us. How is that efficient? AI should be common property.

[–] Uranium3006@kbin.social 24 points 1 year ago

honestly I'd rather open source AI I can run locally. even for something like GPT4 an enterprise-scale operation could afford the hardware

[–] Ubermeisters@lemmy.zip 8 points 1 year ago* (last edited 1 year ago) (1 children)

If it's made from all of us it should be free for all of us.

I'm fine with these researchers going out and scraping the social networks to train models, it's incredibly advantageous to society in general. But it's gotta be crystal clear transparency and it's gotta be limitlessly free to all who want to.

It's the only way that any of this won't result in another massive boundary between the 1% and us pod living grunts. It's already a devisively powerful technology when harnessed adversarially, that power is reduced when everyone has access to it as well.

[–] TehPers 1 points 1 year ago (1 children)

If you look at how much they spend per day (poster quoted $700,000 daily but said unverified), how would it make any sense to provide the service for free? I won't argue for/against releasing the model to the public, since honestly that argument can go both ways and I don't think it would make much of a difference anyway except benefit their competitors (other massive companies).

However, let's assume they did release it publicly, what use would that be for the smaller business/individual? Running these models takes some heavy and very expensive hardware. It's not like buying a rack and building a computer, these models are huge. Realistically, they can't provide that as a free service, they'd fail as a company almost immediately. Most businesses can't afford to run these models themselves, the upfront and maintenance costs would obliterate them. Providing it as a service like they have been means they recoup some of the cost of running the models, while users can actually afford to use these models without needing to maintain the hardware themselves.

[–] Clymene@lemmy.ml 2 points 1 year ago (1 children)

Less than a million dollars a day for everyone who wants to in the whole world to use AI right now? That’s peanuts. A single city bus costs $5-800k to buy. Even if costs goes up to several tens of million a day for access for the whole world that’s incredibly affordable.

It’s crazy that something so useful and so cheap to run can’t be sustained in the current system. This seems like an argument against a market based solution to AI.

[–] TehPers 1 points 1 year ago (1 children)

Less than a million dollars a day for everyone who wants to in the whole world to use AI right now?

You're ignoring the fact that the cost scales with usage. Increasing its availability will also increase the cost, hardware requirements (which can't really scale since there's a shortage), and environmental cost due to power usage.

[–] Clymene@lemmy.ml 2 points 1 year ago (1 children)

No, I am not ignoring that. I specifically said:

Even if costs goes up to several tens of million a day for access for the whole world that’s incredibly affordable.

With how many people are already using AI, it’s frankly mind boggling that they’re only losing $700k a day.

You’re also ignoring the fact that costs don’t scale proportionally with usage. Infrastructure and labor can be amortized over a greater user base. And these services will get cheaper to run per capita as time goes on and technology improves.

Finally, there are positive economic externalities to public AI availability. Imagine the improvements to the economy, education and health if everyone in the world had free access to high quality AI in their native language, no matter how poor or how remote. Some things, like schools, roads and healthcare, are not ideally provisioned under a free market. AI is looking to be another.

[–] TehPers 1 points 1 year ago* (last edited 1 year ago) (1 children)

Finally, there are positive economic externalities to public AI availability.

There are positive economic externalities to public everything availability. We don't live in this kind of world though, someone will always try to claim a larger share due to human nature. That being said, I'm not really interested in arguing about the political feasibility (or lack thereof) of having every resource being public.

With how many people are already using AI, it’s frankly mind boggling that they’re only losing $700k a day.

There are significant throttles in place for people who are using LLMs (at least GPT-based ones), and there's also a cost people pay to use these LLMs. Sure you can go use ChatGPT for free, but the APIs cost real money, they aren't free to use. What you're seeing is the money they lost after all the money they made as well.

You’re also ignoring the fact that costs don’t scale proportionally with usage. Infrastructure and labor can be amortized over a greater user base. And these services will get cheaper to run per capita as time goes on and technology improves.

I don't disagree that the services will get cheaper and that costs don't scale proportionally. You're most likely right - generally speaking, that's the case. What you're missing though is that there is an extreme shortage of components. Scaling in this manner only works if you actually have the means to scale. As things stand, companies are struggling to get their hands on the GPUs needed for inference.

[–] Clymene@lemmy.ml 2 points 1 year ago (1 children)

There are positive economic externalities to public everything availability. We don’t live in this kind of world though, someone will always try to claim a larger share due to human nature.

Saying "Things are inevitably bad because of human nature" is just very weird, since we obviously do have good policies and we try to solve other problems like crime and poverty. It sounds like you already agree that this is good policy? You're just saying it's not politically feasible? OK, sure, we probably don't disagree then.

That being said, I’m not really interested in arguing about the political feasibility (or lack thereof) of having every resource being public.

I am obviously NOT arguing that every resource should be public. This discussion is about AI, which was publicly funded, trained on public data, and is backed by public research. This sleight of hand to make my position sound extreme is, frankly, intellectually dishonest.

there’s also a cost people pay to use these LLMs.

OK, keep the premium subscription going then.

What you’re missing though is that there is an extreme shortage of components.

There's a shortage, but it's not "extreme". ChatGPT is running fine. I can use it anytime I want instantly. You'd be laughed out of the room if you told AI researchers that ChatGPT can't scale because we're running out of GPUS. You seem to be looking for reasons to be against this, but these reasons don't make sense to me, especially since this particular problem would exist whether it's publicly owned or privately owned.

[–] TehPers 1 points 1 year ago

OK, sure, we probably don’t disagree then.

We probably don't here, but like I said I'm not really interested in discussing the political feasibility of it.

I am obviously NOT arguing that every resource should be public. This discussion is about AI, which was publicly funded, trained on public data, and is backed by public research. This sleight of hand to make my position sound extreme is, frankly, intellectually dishonest.

I don't think I ever disagreed that the models themselves should be public, and there are already many publicly available models (although it would be nice if GPT-N were). What I disagree with is the service being free. The service costs a company real money and resources to maintain, just like any other service. If it were free, the only entity that could reasonably run the models is the government, but at this point we might as well also have the government run public git servers, public package registries, etc. Honestly, I'm not sure what impression you expected me to get, considering the claim that a privately run service using privately paid-for resources should be free to the public.

There’s a shortage, but it’s not “extreme”. ChatGPT is running fine. I can use it anytime I want instantly. You’d be laughed out of the room if you told AI researchers that ChatGPT can’t scale because we’re running out of GPUS.

Actually no, I work directly with AI researchers who regularly use LLMs and this is the exact impression I got from them.

[–] Maximilious@kbin.social 29 points 1 year ago (3 children)

Have they asked the engine how they can generate more revenue?

[–] Tigbitties@kbin.social 11 points 1 year ago

$700k/dsy. That's $260 million per year. There are 360 million paid seats of MS360. So they'd have to raise their prices $0.73 per year to cover the cost. No one will notice.

[–] Ubermeisters@lemmy.zip 4 points 1 year ago

It's answer was 42

[–] bahmanm@lemmy.ml 2 points 1 year ago
[–] donuts@kbin.social 25 points 1 year ago (4 children)

AI as a business is already running on fumes, and it's going to become even more expensive once intellectual property law catches up to them. We can only hope that the AI bubble bursting doesn't take the entire market economy down with it...

[–] Taako_Tuesday@lemmy.ca 9 points 1 year ago* (last edited 1 year ago)

At this point there are so many bubbles, the question is which one will burst first

[–] borlax@lemmy.borlax.com 7 points 1 year ago (1 children)

Or we can hope that it does take the entire market economy down with it...

[–] donuts@kbin.social 2 points 1 year ago (1 children)

I mean, I get you, but personally I don't really like the idea of millions of innocent people losing their homes and most of their savings because some fucking dweebs decided to put all of our collective wealth in legally dubious automatic junk "content" generators. I've lived through enough crashes to know that it's never the big guys that get fucked when everything goes tits up, it's us, our parents, our grandparents, etc.

[–] borlax@lemmy.borlax.com 1 points 1 year ago

Yeah status quo is the only reason to not throw caution to the wind and burn the whole thing down. It’s why nothing will ever get better.

[–] mojo@lemm.ee 1 points 1 year ago

Well it doesn't help that ChatGPT is unoptimized as fuck with like 185b parameters for 3.5, and somewhere in the trillions for 4

[–] saud@lemmy.ml 20 points 1 year ago (2 children)

I really don't understand how this is possible when Microsoft has invested billions into OpenAI

[–] Jessper 15 points 1 year ago (1 children)

It isn't. The article is obviously wrong.

[–] Peanutbjelly@sopuli.xyz 13 points 1 year ago

And you are the only voice of reason in this thread.

"Make up shit that makes OpenAI look bad" is like tech article gold right now. The amount of times i am seeing "look what ChatGPT said!!!" As if prompter intention is completely irrelevant to model output.

Objectivity doesn't exist anymore. It's just really popular to talk shit about ai right now.

Like when Altman effectively said "we should only regulate models as big or bigger than ours, we should not regulate small independent or open source models and businesses" to Congress, which was followed by endless articles saying "Sam Altman wants to regulate open source and stamp out smaller competition!"

I have no love for how unopen they've become, but at least align criticisms with reality please.

[–] Uncle_Bagel@midwest.social 8 points 1 year ago

Burning through billions of investors money isnt the same as being profitable. The Silicon Valley gravy train is over, and investors are actually demanding to start seeing returns on their investments.

[–] boyi@lemmy.sdf.org 12 points 1 year ago* (last edited 1 year ago)

Sorry to say, I would take this with grain of salt. Not making profits is part of business model of these pioneering companies. Google, Amazon and Uber (etc) were in the negatives for so many years and they absorbed the losses in order to be the dominant brands where at the end users become dependent on them. At that point they'll start to charge exorbitantly and forcefully add unneeded features that will exert more control upon their users but there's nothing that they can do but pay, for the simple fact that they can't do without them.

[–] miss_brainfart@lemmy.ml 12 points 1 year ago

This is hilarious if true

[–] autotldr@lemmings.world 9 points 1 year ago

This is the best summary I could come up with:


According to the study, the chatbot's responses have worsened despite OpenAI's efforts to ship new features to the tool designed to improve its usability.

Not to mention the amount of money used to procure GPUs from companies like NVIDIA to ensure that things run seamlessly.

Aside from monetary issues, OpenAI is also experiencing a decline in the number of users that leverage its chatbot's offerings.

OpenAI's APIs have increasingly gained the interest of organizations initially opposed to the whole Artificial Intelligence idea and incorporating it into their workflows.

And while OpenAI continues to invest in the venture heavily, Altman has also expressed his concerns over safety measures to ensure that the tool doesn't spiral out of control.

The FTC already launched an investigation into ChatGPT to determine whether the company has broken consumer protection laws.


I'm a bot and I'm open source!

[–] mojo@lemm.ee 9 points 1 year ago

They also didn't design ChatGPT to be power efficient at all, so that's bloating up their operating costs a ton.

[–] roguetrick@kbin.social 9 points 1 year ago

High interest rates baby. I noted this was happening when people were complaining about lowered quality because they were using less resource intensive operations.

[–] deFrisselle@lemmy.sdf.org 9 points 1 year ago

MS will buy it all at the fire sale for cheap then integrate it deeply into Bing, Windows, etc

[–] codepengu1n@feddit.it 8 points 1 year ago (1 children)

If Twitter ran for decades on a loss, so will OpenAI. Worst case scenario they get completely absorbed by MS and have the bill footed by them. Kind of what happened with Youtube.

[–] worfamerryman 7 points 1 year ago (1 children)

One of the reasons tech companies were able to operate for a long time at a loss was due to interest rates being really low.

I think the increased interest rates are the reason for twitter, reddit, and other companies trying to monetize harder.

I could be wrong, I am not an economist. This is just the kind of stuff, I have read in other subs.

[–] codepengu1n@feddit.it 2 points 1 year ago (1 children)

That’s definitely part of the truth. However, something as valuable as OpenAI would still be worth paying for for some tech giant that can afford it.

[–] SmoothIsFast@citizensgaming.com 1 points 1 year ago (1 children)

Maybe before they just decided they could make it bigger without much refinement and now have a completely shit system that's just a glorified chat bot with a high ego to assert its false knowledge.

[–] codepengu1n@feddit.it 1 points 1 year ago (1 children)

I don’t think that’s true. Maybe ChatGPT is a glorified chatbot, but their GPT-4 stuff is pretty good.

Gpt 4 is more convincing without any less hallucinations, it's not better its worse due to that fact.

[–] tourist@community.destinovate.com 7 points 1 year ago (1 children)

How do I approach a VC firm about losing all their money in a big spectacle? Seems to be all the rage these days.

load more comments (1 replies)
[–] wahming@monyet.cc 7 points 1 year ago

It's kinda hilarious how many people are swallowing this article without any salt whatsoever.

[–] veloxy@lemm.ee 6 points 1 year ago

Did they try turning it off and not on again? 🙃

[–] philluminati@lemmy.ml 5 points 1 year ago (1 children)

Uber has never turned a profit for 14 years. I’m guessing it’s windows central who is facing bankruptcy, intellectually speaking.

[–] Topdog@lemmy.ml 5 points 1 year ago
[–] victron@programming.dev 4 points 1 year ago

That was a quick bubble

[–] Semi-Hemi-Demigod@kbin.social 4 points 1 year ago (1 children)

I wonder what song it will sing when they finally pull the plug

[–] 108@kbin.social 1 points 1 year ago
[–] library_napper@monyet.cc 4 points 1 year ago (1 children)

If they go bankrupt, what happens to their IP? Does it suddenly become public domain?

[–] 8ace40@programming.dev 9 points 1 year ago

I think it would be auctioned and sold to the highest bidder.

[–] bahmanm@mastodon.social 2 points 1 year ago* (last edited 1 year ago)

@bahmanm@lemmy.ml @hrefna I know you've been intimately following up on the fate of Haidra & Nivenly. Would love to know your thoughts re 👆