The AI bill that has Big Tech panicked (www.vox.com)
from jeffw@lemmy.world to technology@lemmy.world on 16 Jun 2024 01:41
https://lemmy.world/post/16577150

#technology

threaded - newest

njm1314@lemmy.world on 16 Jun 2024 01:47 next collapse

That’s always a good sign I guess.

CMDR_Horn@lemmy.world on 16 Jun 2024 02:00 collapse

Anytime a fortune500 is against something by saying it’s essentially bad for business, then I’m all for it

Mnemnosyne@sh.itjust.works on 16 Jun 2024 17:52 collapse

They’ve gotten smart enough to use reverse psychology on this kind of thing.

This very much feels like “Only please, Brer Fox, please don’t throw me into the briar patch.”

BallShapedMan@lemmy.world on 16 Jun 2024 02:13 next collapse

This all sounds smart to me. I’ll vote for it given the chance.

Even_Adder@lemmy.dbzer0.com on 16 Jun 2024 04:43 collapse

If it’s the same one from a few months ago, the wording is so vague that only huge companies with legal departments will be able to navigate the compliance maze they’ve set up.

eff.org/…/generative-ai-policy-must-be-precise-ca…

muntedcrocodile@lemm.ee on 16 Jun 2024 02:23 next collapse

Sounds like this is will foss models. This is why all the bug tech companies are pushing ai dangerous narrative they gonna legislate away our freedom for moss models to keep hold of a monopoly. This is how liberty dies with thunderous applause.

deranger@sh.itjust.works on 16 Jun 2024 03:00 next collapse

legislation in the works that mandates that companies that spend more than $100 million on training a “frontier model” in AI — like the in-progress GPT-5 — do safety testing. Otherwise, they would be liable if their AI system leads to a “mass casualty event” or more than $500 million in damages in a single incident or set of closely linked incidents.

Are those models made by companies that would be affected based on the conditions above?

muntedcrocodile@lemm.ee on 16 Jun 2024 07:35 next collapse

I believe some may reach that and it does set a significant limit on capability of foss.

Grimy@lemmy.world on 16 Jun 2024 16:27 collapse

All models are very costly regardless of open source or closed source, but I’m not sure any current model reaches that high. The 100$ million seems to only applies to the cost of computing and not of buying the actual cards.

The legislation is essentially asking that it can’t make nukes or do massive hacking attacking and only asking it of people that definitely have the money to make sure.

It’s actually very level headed compared to what most are pushing for. I can’t even see it affect current gen AI, which are mostly harmless anyways.

Communist@lemmy.ml on 16 Jun 2024 03:21 next collapse

Do foss models really matter? I’m pro foss and think proprietary software should be banned but these weights are essentially a compiled program, we have no idea what they do

muntedcrocodile@lemm.ee on 16 Jun 2024 07:34 collapse

Some are foss in the sence its free. But u do have a point.

Mnemnosyne@sh.itjust.works on 16 Jun 2024 17:57 collapse

Yup, exactly. The only regulation I’d be in favor of for AI is this: if it was trained on data which can be accessed by or was posted by the public, it must be freely available, such that if anything in the training data was posted online in a way anyone can see, then then I have free access to tge AI too.

Basically any other regulation, even if the companies whine publicly, is actually one that benefits them by raising the barrier of entry and making it more expensive for small actors to create AI tools.

just_another_person@lemmy.world on 16 Jun 2024 03:18 next collapse

Or…just don’t use AI.

These dumb shits act like it’s enriching people’s lives. Instead, it’s just making a very specific group of rich people more wealthy.

It’s a fleecing of suckers who think it’s some useful tool to eliminate human workers that cost money.

FaceDeer@fedia.io on 16 Jun 2024 04:13 next collapse

Go ahead and don't use AI if you don't want to. If you think they're truly useless then they'll just go away on their own, right?

I've been finding various AI tools to be very useful to me, personally.

kata1yst@sh.itjust.works on 16 Jun 2024 04:19 next collapse

Agreed. There’s tons of amazing applications that are advancing astrophysics, mathematics, particle physics, pharmacology, oncology, etc etc etc.

It’s a problem of application and efficiency. Both are getting better at a break neck pace.

Kolanaki@yiffit.net on 16 Jun 2024 04:54 next collapse

Do you find the AI features tacked into literally every modern device and application being sold to the general consumer market useful, or do you find a specific niche AI tool meant for specific industry use useful?

admin@lemmy.my-box.dev on 16 Jun 2024 05:35 collapse

Moving goal posts.

Op said Don’t use AI, you’re saying Don’t use AI everywhere.

Kolanaki@yiffit.net on 16 Jun 2024 05:39 collapse

I think it’s pretty clear what type of AI OP was talking about.

webghost0101@sopuli.xyz on 16 Jun 2024 05:50 collapse

I commend you for believing most people know the difference.

Sadly thats not my experience.

Kolanaki@yiffit.net on 16 Jun 2024 06:02 collapse

I just believe people know what they see, and in this instance they are seeing the garbage AI tools being shoved into their phones and computers that give them “information.” They’re not really seeing the good stuff, because the good stuff isn’t being sold to them.

webghost0101@sopuli.xyz on 16 Jun 2024 06:13 collapse

Its true i do worry about shit ai being plastered within our devices but if you cut through the marketing you see a whole mix of machine learning and ai is used under the hood.

Some of these tools like ms paint auto removing backgrounds. And personal assistant like siri talking more fluently does seem like an improvement.

I have no hopes for winfows rewind but even for that we must admit its not actually available yet. Neither is apple ai.

So we are simply assuming that all of the tools they may put in are bad based on some current stupid ideas that are explored.

technocrit@lemmy.dbzer0.com on 16 Jun 2024 15:05 collapse

When all you have is hype, everything looks like intelligence.

RedditWanderer@lemmy.world on 16 Jun 2024 06:56 collapse

Usefulness is one thing, but it costs an astronomical amount of energy.

These companies are trying to make taxpayers pay for their infrastructure by pretending it’s to benefit everyone. It won’t benefit everyone that’s for sure.

evranch@lemmy.ca on 16 Jun 2024 07:21 collapse

It’s possible for local AI models to be very economical on energy, if used for the right tasks.

For example I’m running RapidOCR which uses a modern transformer architecture, and absolutely blows away traditional OCR at capturing data from character displays.

Doesn’t even need a GPU and returns results in under a second on a modern CPU. No preprocessing needed, just feed it an image. This little multimodal transformer is just as much “AI” as bloated general purpose GPTs, but it’s cheap, fast and useful.

RedditWanderer@lemmy.world on 16 Jun 2024 08:19 next collapse

That’s cool and all, but we’re talking about the AI companies that are trying to get valuated at trillions of dollars and want taxpayers to pay for the upgrades to the grid. The sad part is it’s likely going to work

Grimy@lemmy.world on 16 Jun 2024 15:56 next collapse

I’m all for having companies pay for their electricity use and their impact on the grid but that has nothing to do with AI.

Llama took 2 600 mWh to train over 6 months and can run on much less than what’s needed for gaming. ActivisionBlizzard used 86 000 mWh of energy in 2022 for both the datacenters for their games and the development of them. Yet no one in their right mind would suggest to curb stomp gaming to save on energy.

Openai has bigger costs but they run inference, and having them run it actually makes it more efficient, even though I rather open source models you can run on your own machine.

The clear solution is upgrading to a more robust green energy grid, not blocking innovation.

And if we are going to ban things because of their energy use, there are much better candidates than software. A transatlantic flight takes up 500 mWh, so essentially 1000 people flying to Europe and back use up as much energy as the llama model took to train, a model that has been downloaded 3.5 million times in the past month alone on hugging face (only with the official 8b included, and not counting the other sizes or the thousands of finetunes).

auzas_1337@lemmy.zip on 16 Jun 2024 16:17 next collapse

Have you got something to read up on regarding comparisons of energy consumption? Sounds really interesting, but I know close to jack shit about this.

Grimy@lemmy.world on 18 Jun 2024 12:57 collapse

Most big companies publish their energy usage like the two examples above. For the plane bit, I just found multiple people calculating it and coming up with the same number online, so that one might be hot air.

RedditWanderer@lemmy.world on 16 Jun 2024 20:54 collapse

That’s completely besides the point.

Blizzard isn’t asking taxpayers to subsidize them billions “to advance humanity”.

As you say yourself, there are way better models than what is being funded right now, and what is likely to get the monopoly on energy, at our expense.

evranch@lemmy.ca on 16 Jun 2024 16:12 collapse

I’m just stating that “AI” is a broad field. These lightweight and useful transformer models are a direct product of other AI research.

I know what you mean, but simply stating “Don’t use AI” isn’t really valid anymore as soon these ML models will be a common component. There are even libraries and hardware acceleration support for tensor operations on the ESP32-S3.

RedditWanderer@lemmy.world on 16 Jun 2024 20:57 collapse

I didn’t say don’t use AI.

technocrit@lemmy.dbzer0.com on 16 Jun 2024 15:03 collapse

As usual with “AI”, there’s no intelligence involved with OCR. It’s just more data processing / classification being lumped into the hype.

evranch@lemmy.ca on 16 Jun 2024 16:19 collapse

Right, we need to come up with better terms for talking about “AI”. Personally at the moment I’m considering any transformer-type ML system to be part of the category, as you stated none of them are any more “intelligent” than any others. They’re all just a big stack of tensor operations. So if one is AI, they all are.

Remember long ago when “fuzzy logic” was all the hype and considered to be AI? Just a very early form of classifier network but everyone was super excited at the time.

Potatisen@lemmy.world on 16 Jun 2024 06:03 collapse

It’s a positive that it removes jobs. The negative is that society can’t deal with it.

NeoNachtwaechter@lemmy.world on 16 Jun 2024 20:25 collapse

it removes jobs.

They can work at the power plants then. You know, we need so many more power plants, in order to feed our Great and Hungry AI.

/s

NeoNachtwaechter@lemmy.world on 16 Jun 2024 07:16 next collapse

warned that liability for mass casualties caused by AI will destroy the industry

Get real, man!

If liability really can destroy an industry, then this industry should never have existed in the first place.

pastermil@sh.itjust.works on 16 Jun 2024 14:47 next collapse

I guess if AI can destroy humanity, then humanity should never have existed in the first place.

technocrit@lemmy.dbzer0.com on 16 Jun 2024 14:59 collapse

What an inane analogy. “Humanity” is not liable for the actions of legal entities like corporations. Should we all be punished for the misdeeds of boeing? It’s probably only a few people who are directly responsible. Don’t shill for them.

atrielienz@lemmy.world on 16 Jun 2024 18:44 collapse

Did they need a slash s for this? Did they? Because people like you make me believe they needed a slash s. Like. Obviously this was a sarcastic comment because the original comment they responded to was horribly fallible. There are whole industries built on the idea that an industry can be destroyed by liability. It’s literally why we have liability insurance. So when someone responds to that comment with an equally fallible statement that is clearly meant to be sarcastic we just ignore that because we feel that their statement is wrong? What even is this.

fluxc0@lemmy.world on 20 Jun 2024 23:40 collapse

with the crazy shit on the internet nowadays, you have to assume the worst at all times. not everyone is good at sarcasm, even when it’s obvious.

rottingleaf@lemmy.zip on 16 Jun 2024 17:57 next collapse

However I dislike this, in some sense we (as in Web users) started this idea that tech should be free from liability.

Then vultures came and try to both make it all work for them and at the same time be free from liability.

NeoNachtwaechter@lemmy.world on 16 Jun 2024 18:10 next collapse

tech should be free from liability.

I call that a childish idea.

rottingleaf@lemmy.zip on 16 Jun 2024 18:50 collapse

Yeah, I call everything absolute a childish idea.

Just like everything else, laws work when there is an alternative. When that ends, they are abused more and more by bad people.

pdxfed@lemmy.world on 16 Jun 2024 21:05 collapse

Just like the letter I got yesterday from an ISP I haven’t done business with in 4 years, letting me know my birthdate and SSN were compromised. Why did they even maintain it if they didn’t have a need for it? Also, why did an ISP need that in the first place…

They offered 1 year of credit monitoring. Lol. I’ll wait for the class action.

grrgyle@slrpnk.net on 17 Jun 2024 01:20 collapse

When you move fast and break things, but then have to pay to fix the things you broke 🥺

NeoNachtwaechter@lemmy.world on 17 Jun 2024 03:12 collapse

Let’s hope that it can be fixed simply with money

lemmy.world/post/16613815

squirrel@lemmy.blahaj.zone on 16 Jun 2024 10:39 next collapse

AI execs: Our AIs are going to be so powerful. More powerful than anything. Soon they could be able to destroy humanity!

Governments: Well, then we better regulate that shit and make sure that doesn’t happen…

AI execs: Nooooo! We did not mean it that way!

General_Effort@lemmy.world on 16 Jun 2024 22:47 next collapse

Why is she claiming that the bill is about liability?

Sabata11792@ani.social on 17 Jun 2024 03:30 next collapse

“frontier model” in AI — like the in-progress GPT-5 — do safety testing. Otherwise, they would be liable if their AI system leads to a “mass casualty event” or more than $500 million in damages in a single incident or set of closely linked incidents.

If your Ai takes over the world and nukes half of it, you will have to pay a fine.

dreikelvin@lemmy.world on 17 Jun 2024 08:08 collapse

what enriches lives:

  1. solving world hunger
  2. doing the taxes and other boring stuff
  3. translation
  4. replacing corrupt governments
  5. cheaper living

what we use AI for instead:

  1. making society, artists, already poor people poorer
  2. making life more complicated thanks to increased joblessness
  3. causing more polarisation and conflict
  4. helping corrupt governments
  5. more expensive living

why invent an AI that eats ice cream for you when instead, it should do the dishes and pass the butter?