Wikipedia is under assault: rogue users keep posting AI generated nonsense

[Cal Jeffrey]

This is why we can't have nice things: Wikipedia is in the middle of an editing crisis at the moment, thanks to AI. People have started flooding the website with nonsensical information dreamed up by large language models like ChatGPT. But honestly, who didn't see this coming?

Correction (10/11/24): Lebleu reached out to clarify a few detailed regarding the motivations behing the user posting AI-generated materiel and the examples mentioned. These changes were applied the original text.

Edited article continues below.

Wikipedia has a new initiative called WikiProject AI Cleanup. It is a task force of volunteers currently combing through Wikipedia articles, editing or removing false information that appears to have been posted by people using generative AI.

Ilyas Lebleu, a founding member of the cleanup crew, told 404 Media that the crisis began when Wikipedia editors and users began seeing passages that were unmistakably written by a chatbot of some kind. The team confirmed the theory by recreating some passages using ChatGPT.

1: AI is hallucinating events, historical figures, entire concepts on Wikipedia

2: a task force of Wikipedia editors is detecting and deleting this stuff https://t.co/PlfzVCZd4P

– Jason Koebler (@jason_koebler) October 9, 2024

"A few of us had noticed the prevalence of unnatural writing that showed clear signs of being AI-generated, and we managed to replicate similar 'styles' using ChatGPT," said Lebleu. "Discovering some common AI catchphrases allowed us to quickly spot some of the most egregious examples of generated articles, which we quickly wanted to formalize into an organized project to compile our findings and techniques."

For example, There is one article about an Ottoman fortress built in the 1400s called "Amberlisihar." The 2,000-word article details the landmark's location and construction. Unfortunately, Amberlisihar does not exist, and all the information about it is a complete hallucination peppered with enough factual information to lend it some credibility. The team identified it as a hoax and deleted it.

 

 

 

 

View this post on Instagram

 

 

 

 

 

 

 

 

 

 

 

A post shared by TechSpot (@thisistechspot)

As for why this is happening, the cleanup crew thinks there are three primary reasons.

"[The] main reasons that motivate editors to add AI-generated content: self-promotion, deliberate hoaxing, and being misinformed into thinking that the generated content is accurate and constructive," Lebleu said.

However, let's be honest – there are two factors that are the main contributors to this monkey business. First is an inherent problem with Wikipedia's model – anyone can be an editor on the platform. Many universities do not accept students turning in papers that cite Wikipedia for this exact reason.

The second is simply that the internet ruins everything. We've seen this time and again, particularly with AI applications. Remember Tay, Microsoft's Twitter bot that got pulled in less than 24 hours when it began posting vulgar and racist tweets? More modern AI applications are just as susceptible to abuse as we have seen with deepfakes, ridiculous AI-generated shovelware books on Kindle, and other shenanigans.

Anytime the public is allowed virtually unrestricted access to something, you can expect a small percentage of users to abuse it. When we are talking about 100 people, it might not be a big deal, but when it's millions, you are going to have a problem. Sometimes, it's for illicit gain. Other times, it's just because they can. Such is the case with Wikipedia's current predicament.

Permalink to story:

Wikipedia is under assault: rogue users keep posting AI generated nonsense

 

[NumberNine]

The second reason is simply that the internet ruins everything.

Incorrect. The way people use the internet ruins everything.
Let's not pretend this is some passive phenomenon that happens on its own.

 

[kiwigraeme]

I think it's it in the first sentence, why we can't have nice things.
Imagine in your city how many great facilities, beautification projects you could have if not just kids , adults take pleasure in stealing , defacing and destroying it

Yes trolls get a easier ride, can be anonymous except where they brag

There is no simple answer. Folks are even toxic/downers in work places taking fun it killing peoples joy and initiatives, because to them its nothing or I don't like that so everyone must not like it. I'm doing the office a favour by chucking it in the bin, while everyone is away

Ignoring the angry, immature, feel outside of their community and ostracised.

There always been those who want to make it about themselves. Ignoring people like myself who think I have better than average critical skills, we don't go controlling and demanding our opinions onto others. We know others thing different, there is no one right answer etc

This kind of reminds me of levels of self-actualisations, if I take photos of dead people in car crash I can post it immediately , I will will liked and and doing a great service. yet someone with a greater understanding would realise this is not OK and not good for yourself

To the topic at hand, it someone gets a great prank article into wikipedia that is pretty harmless, on the whole it's entertaining, some comedians have pranked main stream media creating completely bogus stories.

But this just seems mindless and destructive for sake of being the main character
Problem is even AI checkers can be set to fail if fake sources, stories put out on WWW

Thing is people like this who want to be free, Me Me Me, socialism is bad etc. Are often stupid hypocrites if societies benefits are taken away from them . That's why at least base jumpers have some respect , they don't whine when one of them dies. Most of the trolls on the internet are nothing like this and can't handle the jandal.
Narcistic people doing bad stuff , can't handle even constructive feedback

 

[Axeia]

It's not just just Wikipedia. There's tons of sites filled with AI generated nonsense.
That nonsense will be fed back into AI again resulting in the nonsense being fed back into newer models.

We've reached the point where we can't trust any information that isn't from before the advent of AI.

 

[captaincranky]

IMO, our entire "highly advanced species", is the biggest sh*t stain on the planet evolution has misguidedly ever produced.

As an "apex predator", we've pretty much selfishly taken most of what other creatures need to barely exist. The odds in our, "favor", being 8,000,000,000 to maybe 100 at most. Our predation extends to the planet itself.

Maybe 1/10% (at most) of all living humans contribute to "moving us forward". Whereas, the other 99 9/10% do virtually nothing but, "sh*t in our own nest".

As to defining, "moving us forward, a great many of the things we enjoy and depend on, are offshoots, adaptations or implementations of weapons originally designed for warfare. Take the jet engine as an example.

In other words, "give the ape anything", and he'll quickly find a way to misuse or abuse it. That includes everything from a spray can of paint, to an advanced AI software.

 

[human7]

"Many universities do not accept students turning in papers that cite Wikipedia for this exact reason."

While true, it's not the main reason. Wikipedia, like other encyclopedias, is a tertiary source. At the university level, encyclopedias simply aren't deep or focused enough to cite; papers should be citing primary and secondary sources.

 

[MarkHughes]

I don't often visit Youtube but the other day I saw a video there and every comment was a rewording of the same basic sentence, And there where dozens and dozens of them. Rendering the comments section useless (Not necessarily a bad thing in this case). I assume all sites will suffer from this eventually.

 

[MarcusNumb]

At this rate, soon enough students or everyone will have to abandon internet as a source for information, we'll all have to read actual paper books to learn anything.

 

[FaTaL]

Some people are sick and bored. a dangerous combination.

 

[captaincranky]

Some people are sick and bored. a dangerous combination.

That a correct and acute observation if somewhat trite However, it doesn't explain cause and effect relationship for the uptick in mental health issues.

For example, could rap music be responsible for the gun violence taking place in our inner cities? The lyrics, (if you can call them such), espouse violence and killing, especially with a "nine". Kudz go out and kill one another with guns. Could rap be responsible? Of course not. After all, the money men in the music industry who likely paid for studies about this, say that's not possible.

Could Facebook be the cause of teenage suicide We'll just have to keep our kidz off the web, or not raise such snowflakes.

You'll have to excuse me, it's past my dinner time. Think I'll plop down a couple of yummy Tide Pods.

 

[Axeia]

At this rate, soon enough students or everyone will have to abandon internet as a source for information, we'll all have to read actual paper books to learn anything.

And you think that there aren't 'authors' generating books full of AI generated nonsense to get printed and make a buck?
Both internet and books are simply going to be don't trust anything from before ~2022.

It would be ironic if what is supposedly the biggest innovation since the internet is what largely undoes the internet and ultimately might even stunt our progress for a bit.

 

[MarcusNumb]

And you think that there aren't 'authors' generating books full of AI generated nonsense to get printed and make a buck?
Both internet and books are simply going to be don't trust anything from before ~2022.

It would be ironic if what is supposedly the biggest innovation since the internet is what largely undoes the internet and ultimately might even stunt our progress for a bit.

Well then we are doomed. It's ironic how the information era will be f***** by the AI info generated xD

 

[mgwerner]

And you think that there aren't 'authors' generating books full of AI generated nonsense to get printed and make a buck?
Both internet and books are simply going to be don't trust anything from before ~2022.

It would be ironic if what is supposedly the biggest innovation since the internet is what largely undoes the internet and ultimately might even stunt our progress for a bit.

This is why I will never get rid of my 1986 set of Encyclopaedia Britannica, with the decade of updates.