this post was submitted on 28 Jun 2026
362 points (100.0% liked)

Not The Onion

21948 readers
550 users here now

Welcome

We're not The Onion! Not affiliated with them in any way! Not operated by them in any way! All the news here is real!

The Rules

Posts must be:

  1. Links to news stories from...
  2. ...credible sources, with...
  3. ...their original headlines, that...
  4. ...would make people who see the headline think, “That has got to be a story from The Onion, America’s Finest News Source.”

Please also avoid duplicates.

Comments and post content must abide by the server rules for Lemmy.world and generally abstain from trollish, bigoted, ableist, or otherwise disruptive behavior that makes this community less fun for everyone.

And that’s basically it!

founded 3 years ago
MODERATORS
 

Alice* says she feels “not in the slightest” guilty about using ChatGPT to complete training tasks, saying it is easy to get away with as long as you instruct chatbots to avoid the usual telltale signs of AI output, like a preponderance of em-dashes. “It’s only the sloppiest of users that get caught,” she says. “Anyone with a modicum of awareness around AI hallmarks can tell their output not to use them, and at that point what are you going to do?”

Another worker, Bob*, worked for a training platform called Outlier. Initially, he was tasked with AI training, which he says he illicitly used AI for, and was then promoted to a leadership role where part of his job was to catch others doing the same thing.

all 31 comments
sorted by: hot top controversial new old
[–] SGGeorwell@lemmy.world 77 points 2 weeks ago (1 children)

Maybe if we force feed the Sloppies their own slop, they’ll slop themselves to death! Fight slop with slop.

[–] Infrapink@thebrainbin.org 19 points 2 weeks ago (1 children)

Yup. LLMs won't go away unless and until they collapse in on themselves, so I say let's accelerate it.

[–] some_designer_dude@lemmy.world 6 points 2 weeks ago (3 children)

For anyone else reading, this plan is pretty naive. They’d just cut back to an earlier, un-poisoned model and be far more careful about who trains them, and how. Even if the general public briefly loses trust in the whole concept, that’s not where the money is anyway.

[–] CADmonkey@lemmy.world 4 points 2 weeks ago

That sounds expensive.

[–] Epp@lemmus.org 4 points 2 weeks ago

Shhh...You're ruining the mob's fantasy!

[–] Bane_Killgrind@lemmy.dbzer0.com 2 points 2 weeks ago

There aren't any models that aren't poisoned somehow

[–] laranis@lemmy.zip 66 points 2 weeks ago (2 children)
[–] floquant@lemmy.dbzer0.com 50 points 2 weeks ago (1 children)

Model collapse is the technical term

[–] Infrapink@thebrainbin.org 29 points 2 weeks ago

No, model collapse is the result. The process is more like if the Human Centipede was an ouroboros and also a vampiric fractal Markov chain.

[–] Zarobi@aussie.zone 11 points 2 weeks ago

We actually made an ouroboros. Hooray

[–] Chaunticleer@lemmy.dbzer0.com 26 points 2 weeks ago

Lol. Lmao even

[–] Zombie@feddit.uk 25 points 2 weeks ago (2 children)

“Management vacillated between light tolerance to outright banning,” says Bob. Workers at Outlier would be tracked with a tool called Hubstaff which takes screenshots of their desktop at random intervals to ensure they are really doing tasks as ordered. Bob would look for evidence of AI models in those screenshots.

[–] errer@lemmy.world 15 points 2 weeks ago (1 children)

Just point a webcam at the screen of the computer, OCR the text, feed that to an LLM, and have it type via a KVM. Limit the typing speed to human level typing. I bet there are farms in French Guiana and similar places where they do exactly this.

[–] RedstoneValley@sh.itjust.works 12 points 2 weeks ago* (last edited 2 weeks ago)

You don't need the OCR step... a lot of LLM models come with "vision" They can transcribe your shaky mobile phone pictures and do their slop thing in one step. You could even wire up an output tool MCP style that feeds the slop back via keyboard USB. The perfect Slop Centipede

[–] saltesc@lemmy.world 11 points 2 weeks ago* (last edited 2 weeks ago) (1 children)

Those screenshots and recordings are a treasure trove if ever/when they are externally compromised.

[–] foggy@lemmy.world 18 points 2 weeks ago* (last edited 2 weeks ago) (1 children)

Recently was in a CTF where finding the stored screenshots from copilots recall was the path to the flag. A screenshot of a notepad doc that had credentials for the admin user of a compromisable service.

Doc wasn't saved. Copilot didn't care.

Yes, this means if you ever copy/paste a password in clear text temporarily on a machine with recall, and your user is compromised, that password could still possibly be found.

[–] quick_snail@feddit.nl 1 points 2 weeks ago

God forbid you have to edit a config file on ssh

Ups the model collapsed.

[–] BehindetheClouds@reddthat.com 14 points 2 weeks ago (1 children)

Everything's a copy of a copy of a copy.

[–] quick_snail@feddit.nl 3 points 2 weeks ago

God I wish I could sleep

[–] FlashMobOfOne@lemmy.world 11 points 2 weeks ago

We can tell. This method definitely reflects the quality (or lack thereof) in the output.

[–] collapse_already@lemmy.ml 10 points 2 weeks ago

Don't worry, the tech bros sold the gullible execs on no more employees, just profits and so our entire economy is now based on completely unrealistic profit assumptions and circular lending dependencies that will collapse when the bubble inevitably bursts. Somehow this will harm the poor the most.

Maybe we can kick the can down the road so biosphere collapse gets us first.

[–] AlteredEgo@lemmy.ml 10 points 2 weeks ago

These workers are normally employed by third parties, rather than AI companies directly, and are often working without full-time contracts and for low pay. That can incentivise them to take shortcuts

Capitalism and the free market is so efficient!

[–] DarkCloud@lemmy.world 7 points 2 weeks ago

They're great at averaging.

Okay at summarizing...

...and not good for much else. But if you love slop! Boy howdy do AI companies have what you need.

[–] Pappabosley@lemmy.world 6 points 2 weeks ago (2 children)

Reminds me of the movie Multiplicity, clones of clones just getting dumber

[–] docandersonn@literature.cafe 2 points 2 weeks ago

Gary? Gary!

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

To be fair...

I do like pizza

[–] Bane_Killgrind@lemmy.dbzer0.com 5 points 2 weeks ago

Throwing themselves upon the wheels

Nice

[–] ReCursing@feddit.uk 2 points 2 weeks ago (1 children)
[–] DarkCloud@lemmy.world 8 points 2 weeks ago

Shitularity.