this post was submitted on 07 Oct 2025
695 points (98.6% liked)

Technology

76041 readers
2647 users here now

This is a most excellent place for technology news and articles.


Our Rules


  1. Follow the lemmy.world rules.
  2. Only tech related news or articles.
  3. Be excellent to each other!
  4. Mod approved content bots can post up to 10 articles per day.
  5. Threads asking for personal tech support may be deleted.
  6. Politics threads may be removed.
  7. No memes allowed as posts, OK to post as comments.
  8. Only approved bots from the list below, this includes using AI responses and summaries. To ask if your bot can be added please contact a mod.
  9. Check for duplicates before posting, duplicates may be removed
  10. Accounts 7 days and younger will have their posts automatically removed.

Approved Bots


founded 2 years ago
MODERATORS
(page 2) 50 comments
sorted by: hot top controversial new old
[–] Buffalox@lemmy.world 72 points 1 week ago* (last edited 1 week ago) (4 children)

I think the fad will die down a bit, when companies figure out that AI will be more likely than humans to make very expensive mistakes that the company has to compensate, and saying it was the AI is not a valid cop out.
I foresee companies will go bankrupt on that account.

It doesn't help to save $100k on cutting away an employee, if the AI causes damages for 10 or 100 times that amount.

[–] shalafi@lemmy.world 23 points 1 week ago (1 children)

When the bubble bursts, whoever is left standing is going to have to jack prices through the roof to put so much as a dent in their outlay. Their outlay so far. Can't see many companies hanging in there at that point.

[–] BanMe@lemmy.world 9 points 1 week ago (1 children)

Not if the IP is purchased by another company leaving the original saddled with the debt, or spun off so the parent company can rebuy it thusly, or the government bails them out, or buys it to be the State AI too, or a bunch of other scenarios in this dark new world ahead.

[–] TronBronson@lemmy.world 2 points 1 week ago

That’s my favorite part. All the stolen IP.

[–] a4ng3l@lemmy.world 8 points 1 week ago

I put my money on AI act here in Europe and the willingness of local authorities to make a few examples. That would help bringing some accountability here and there and stir a bit the pot. Eventually, as AI commodities, it will be less in the light. That will also help.

[–] simplejack@lemmy.world 3 points 1 week ago (2 children)

Agreed, but I do think that some jobs are just going to be gone.

For example, low level CS agents. I worked for a company that replaced that first line of CS defense with a bot, and the end-of-call customer satisfaction scores went up.

I can think of a few other things in my company that had a similar outcome. If the role is gone, and the customers and employees are being served even better than when they had that support role, that role ain’t coming back.

[–] Buffalox@lemmy.world 5 points 1 week ago (2 children)

I'm pretty sure that even consumer services is an area where I saw a computer made an expensive mistake, promising the customer something very expensive, and a court decided the company had to honor the agreement the AI made. But I can't find the story, because I'm flooded with product placement articles about how wonderful AI is at saving cost in CS.
But yes CS is absolutely an area where AI is massively pushed.

[–] Peffse@lemmy.world 5 points 1 week ago (1 children)

Not sure if it's the one you are referring to, but AI gave discounts on flights.

load more comments (1 replies)
[–] architect@thelemmy.club 3 points 1 week ago

The court honored it for now. I expect the future it will be your problem.

Oh but the EU?

Once they are done with North America the EU will be a non issue for them.

load more comments (1 replies)
[–] jaybone@lemmy.zip 40 points 1 week ago (2 children)

LOLLLLLLLL that’s like a third of the US population. Probably half of the number currently employed. There’s no way in hell this useless garbage will take 1/3 to 1/2 of all jobs. Companies that do this will go out of business fast.

[–] skisnow@lemmy.ca 18 points 1 week ago (4 children)

You can tell how competent someone is at something by how good they think AI is at that thing.

load more comments (4 replies)
load more comments (1 replies)
[–] Zephorah@discuss.online 34 points 1 week ago (2 children)

Thus demonstrating the crux of the issue.

I was just looking for a name of a historical figure associated with the Declaration of Independence but not involved in the writing of it. Elizabeth Powel. Once I knew the name, I went through the ai to see how fast they’d get it. Duck.ai confidently gave me 9 different names, including people who were born on 1776 or soon thereafter and could not have been historically involved in any of it. I even said not married to any of the writers and kept getting Abagail Adams and the journalist, Goddard. It was continually distracted by “prominent woman” and would give Elizabeth Cady Stanton instead. Twice.

Finally, I gave the ai a portrait. It took the ai three tries to get the name from the portrait, and the portrait is the most used one under the images tab.

It was very sad. I strongly encourage everyone to test the ai. Easy to grab wikis that would be top of the search anyway are making the ai look good.

[–] merc@sh.itjust.works 22 points 1 week ago

If you understand how LLMs work, that's not surprising.

LLMs generate a sequence of words that makes sense in that context. It's trained on trillions(?) of words from books, Wikipedia, etc. In most of the training material, when someone asks "what's the name of the person who did X?" there's an answer, and that answer isn't "I have no fucking clue".

Now, if it were trained on a whole new corpus of data that had "I have no fucking clue" a lot more often, it would see that as a reasonable thing to print sometimes so you'd get that answer a lot more often. However, it doesn't actually understand anything. It just generates sequences of believable words. So, it wouldn't generate "I have no fucking clue" when it doesn't know, it would just generate it occasionally when it seemed like it was an appropriate time. So, you'd ask "Who was the first president of the USA?" and it would sometimes say "I have no fucking clue" because that's sometimes what the training data says a response might look like when someone asks a question of that form.

[–] Buffalox@lemmy.world 6 points 1 week ago (1 children)

LOL Maybe AI will be the next big job creator. The AI solves a task super fast, but a human has to sort out the mistakes, and spend twice the time doing that, than it would have taken to just do it yourself.

[–] DarkDarkHouse@lemmy.sdf.org 4 points 1 week ago

This what's happening in computer programming. The booming subfield is apparently slop cleaners.

[–] Grandwolf319@sh.itjust.works 32 points 1 week ago (2 children)

If you have a job that you can be confidently wrong without any self awareness after the fact, then yeah I guess.

But I can’t think of many jobs like that except something that is mostly just politics.

[–] Blackfeathr@lemmy.world 25 points 1 week ago (1 children)

Don't forget the vast majority of CEOs.

[–] thisbenzingring@lemmy.sdf.org 7 points 1 week ago

IMO AI would probably do the job of CEO better than a human. It wouldn't be as greedy and would be happy with any growth while being humble enough to make decisions that might be personally embarrassing

[–] WanderingThoughts@europe.pub 2 points 1 week ago

Spam and astroturfing mostly.

[–] Buffalox@lemmy.world 26 points 1 week ago (1 children)

And over the next 50 years it will take 485 million jobs, and the unemployment rate will be 235%.

[–] architect@thelemmy.club 4 points 1 week ago (1 children)
[–] popekingjoe@lemmy.world 5 points 1 week ago

Here's hoping!

[–] thisbenzingring@lemmy.sdf.org 25 points 1 week ago (5 children)

funny... i expected IT workers to be in that list but we're not. AI couldn't do my job but it could be my boss and that frightens me.

[–] BanMe@lemmy.world 9 points 1 week ago

I drove Amazon Flex during Covid, having an AI as your boss is deeply and perpetually unsettling but ultimately doable! Just do what the push notification tells you to do. If you want to say something to your boss, use the feedback form on the corporate website. So simple.

[–] sexy_peach@feddit.org 3 points 1 week ago (1 children)
[–] thisbenzingring@lemmy.sdf.org 3 points 1 week ago

what don't I do.... some days... I tell you. My job is Systems Administrator

load more comments (3 replies)
[–] MonkderVierte@lemmy.zip 23 points 1 week ago* (last edited 1 week ago) (4 children)

Stop calling LLM AI. It creates false expectations.

load more comments (4 replies)
[–] phutatorius@lemmy.zip 20 points 1 week ago

Just look at who's in charge of the Senate, and ask yourself if they are to be trusted to do anything but lie, steal and carry out witch hunts.

As for LLMs, unless driving contact-centre customer satisfaction scores even further through the floor counts as an achievement, so far, all there's been has been a vast volume of hype and wasted energy, and very little to show for it, except for some highly constrained point solutions which aren't significant enough to make economic impact. Even then, the ROI is questionable.

[–] tidderuuf@lemmy.world 19 points 1 week ago (1 children)

Knowing the way our country is going I would expect in the end workers will have to pay an AI tax on their income and most workers will start working 50 hours a week.

[–] Buffalox@lemmy.world 2 points 1 week ago

I like you optimism that it won't be worse than that. 😋

[–] Smoogs@lemmy.world 15 points 1 week ago

Yes but got forbid those jobs be stolen by another country. Can’t have that.

[–] _stranger_@lemmy.world 12 points 1 week ago* (last edited 1 week ago)

So they want to keep them terrified of losing their shitty, barely functioning status quo.

The reality is that these are the numbers the Republicans want , because it's the numbers their billionaire owners want. ChatGPT is just accidentally letting us know how they've poisoned the models.

[–] Kyle_The_G@lemmy.world 4 points 1 week ago

and then 115 million will be needed to unwind the half-assed implementation and inevitable damage.

[–] tal@olio.cafe 4 points 1 week ago* (last edited 1 week ago)

I wouldn't put it entirely outside the realm of possibility, but I think that that's probably unlikely.

The entire US only has about 161 million people working at the moment. In order for a 97 million shift to happen, you'd have to manage to transition most human-done work in the US to machines, using one particular technology, in 10 years.

Is that technically possible? I mean, theoretically.

I'm pretty sure that to do something like that, you'd need AGI. Then you'd need to build systems that leveraged it. Then you'd need to get it deployed.

What we have today is most-certainly not AGI. And I suspect that we're still some ways from developing AGI. So we aren't even at Step 1 on that three-part process, and I would not at all be surprised if AGI is a gradual development process, rather than a "Eureka" moment.

[–] DamnianWayne@lemmy.world 1 points 1 week ago

Well my AI says it will take 96 or 98 million jobs, depending on what you want it say and only for $5,000.

[–] CosmoNova@lemmy.world 1 points 1 week ago* (last edited 1 week ago) (1 children)

Why even post this here? This is politics BS that‘s used as a diversion from the Epstein files and the government shut down which again only happened so they don‘t vote on the Epstein files.

[–] TronBronson@lemmy.world 1 points 1 week ago (2 children)

The epstien files is a distraction from dismantling our constitutional law. What laws are you going to try the pedos under? Which courts do you plan on using? You see where I’m going with this? We all know who’s on the list who’s gonna hold them accountable? No one, thus it’s a stupid distraction.

load more comments (2 replies)
load more comments
view more: ‹ prev next ›