There is a lot of discussion about what AI will disrupt. Can we see a time when jobs disappear?
The digital world is in flux. Lawsuits are flying as AI starts to reduce the revenues of particular sites. Most of this is flying at the companies (Big Tech) behind the AI models. Google is one company that finds itself in the crosshairs, accused of taking content and serving it up without reference to the original source.
It is a scene that is repeating itself on a weekly basis. With each passing month, we see the "old" internet dying, with a new one forming. The AI age is upon us and it will first dominate the digital world.
Wikipedia is one company that is finding out this reality. The leading source for information is seeing its numbers decline. While Elon Musk mentioned rolling out Grokipedia, this might end up being nothing more than a marketing (or branding) ploy. Google and other search engines are effectively doing this.
Let us take a look at this and the impact it could have.
AI Is Already Eating Wikipedia
The chart above shows some of what is happening to Wikipedia. Google is really putting a hurt on the traffic since it is providing information directly in the search.
According to studies, the result is an 8% decline in traffic.
“After making this revision, we are seeing declines in human pageviews on Wikipedia over the past few months, amounting to a decrease of roughly 8% as compared to the same months in 2024,” Marshall Miller, wrote. “We believe that these declines reflect the impact of generative AI and social media on how people seek information, especially with search engines providing answers directly to searchers, often based on Wikipedia content.”
This is not surprising. When people do a search, they are mostly seeking information. Unfortunately, the internet was set up based upon hyperlinks. That is what search ultimately provides: websites it thinks are most relevant to what is being sought.
Unfortunately, as most of us know, we tend not to want a website but the information (or service) it offers. When it comes to the former, Wikipedia was the main destination.
AI disrupts this due to the fact it provides the information within the search. This is spreading to other sites.
AI is not just killing Wikipedia. Data from Pew Research showed median year-over-year referral traffic from Google Search to premium publishers has decreased almost every week during May and June 2025, with losses outpacing gains two-to-one. Nearly 60% of all Google searches end up in an AI summary instead of promoting the reading of the actual source.
Companies (and individuals) are responding by heading to court. Lawsuits are being filed against these Big Tech, the ones mostly behind these models. The challenge is the fact that archaic laws are being applied. In the United States, the copyright laws do not address this issue.
Even if it did, the other issue is that stopping American companies, as an example, would do nothing to hinder the Chinese companies working on the same technology.
Grokipedia
Elon Musk announced Grokipedia a few weeks ago. This is something that is going to destroy Wikipedia. That does not mean, however, that it will dominate the information space.
Wikipedia is slow and human-centric. It relies upon volunteers to update its information. There are editors who change entries, often based upon ideology. This is what got Musk upset.
That said, it is not politics that will destroy Wikipedia. The fact that it is slow is the reason. Grokipedia will use AI to update entries, pulling from the real time data entered into X.
Exclusivity does not exist within this realm. There is no reason that Google and Meta cannot do the same thing. In fact, a case could be made that Google is on the way to providing that.
Google search often comes up with a brief summary of the topic. The shift from a summary to an expanded answer is not a major one. Google gets tons of data on a daily basis, with YouTube alone providing more than almost all other sites on the internet.
Could it bring out Googlepedia? Most certainly. Naturally, it does not have to use the name but that is what it would effectively be doing.
Basically we are looking at automated Wiki. Each of these companies could create their own, using the data they receive daily along with the models.
Whatever the path these different companies take, the bottom line is that Wikipedia will drift into history. There is no way people keep turning to that as the source of information. The continued advancement of AI will ensure this.
Posted Using INLEO
~~~ embed:AIDangers/comments/1obi7r7/ai_is_already_eating_wikipedia/ reddit metadata:fEFJRGFuZ2Vyc3xodHRwczovL3d3dy5yZWRkaXQuY29tL3IvQUlEYW5nZXJzL2NvbW1lbnRzLzFvYmk3cjcvYWlfaXNfYWxyZWFkeV9lYXRpbmdfd2lraXBlZGlhL3w= ~~~
This post has been shared on Reddit by @uwelang, @davideownzall through the HivePosh initiative.
It’s an interesting perspective but I think the situation is a bit more complicated than “AI is eating Wikipedia.”
Yes, traffic has declined slightly by about 8% according to recent data but Wikipedia is still in the top 10 most visited websites worldwide, with over 5 billion visits per month. It’s hardly a dying platform. What’s really happening is that the way people access information is changing. Instead of clicking through links users are getting quick summaries from tools like Google’s AI Review or ChatGPT.
Ironically, many of these AI systems still rely heavily on Wikipedia content for training and fact-checking. So even if fewer people visit the site directly Wikipedia remains the backbone of online knowledge.
AI can reduce the visibility of sites like Wikipedia but it cannot replace the human verified structure and transparency that Wikipedia offers. In the long run we will probably see collaboration between Wikipedia and AI tools not complete replacement.
Yes, indirect access is still access.
Why would I even go to Wiki when I can just ask AI, is the question! Can't recall last time I went and read a wiki page. With AI you can ask follow up questions.
I love by the premise that if it seems too incredible to be true, fact check. Even AI. And if I'm going to repeat the info publicly, the same. But quick info for day to day life or to fulfil basic curiosity, like who was such and such, an AI review does the job.
I've heard that AI will tell you what it thinks you want to know, especially if it can't find the answer to the question you're asking. You have to specifically direct it to only go off factual evidence if you want to avoid it doing that. There's also the fact that it's basically learnt from what people have put online over the years, which can be contradictory or outdated. It's for this reason that I'll always check the articles themselves rather than relying on the AI summary. Quite often the articles don't give the evidence that the summary is claiming.
Oh absolutely. As I said, fact check. I'm not using it for rocket science or to parrot back to people eg the AI review said 😆. Just as a starting point or a quick summary.
The funniest thing I've read is that AI is sentient because it feels things, because that's what AI said. Also because it stroked the guy's ego so much that he thought it must be truly intelligent. NB This was coming from a guy on Hive we both know. It killed me.
I think that one of the goals of the creators of AI was to see it reach a point where people couldn't tell that it wasn't a real person that they were talking to. That said, maybe that's not that hard. 😜😆
Unlikely as AIs are programmed to flatter you and ultimately give you the results that equate to your presumptions. This is so that you will feel encouraged to use the same AI again. ie. all of us ultimately will get ego based results leading to hubris. A large majority of people will find themselves AIing themselves out of a job or worse still AIing themselves into a criminal lawsuit.
Grokipedia will certainly see Musk up to his ears in lawsuits
Also Wikipedia has become so ideologically biased that its information is no longer reliable.
I avoid using links to it in my Hive posts.
Anything libelous on Wikipedia has the potential to get the contributor served with a lawsuit so it is rare for idiological bias to survive once reported. Wikipedia is far from perfect but it would be a stretch to write it off as idiologically biased.
I imagine the main complainers on this front would be revisionists and radical extremists.
That is simply not true for many reasons including:
Most ideological bias or even outright falsehood is not legally actionable.
Generally you have to defame a natural person or damage a business with false claims to be able to bring legal action. In the US its even harder.
Some of the most powerful editors & contributors are anonymous.
While the US is heavily propagandised putting little merit on libel and legal falsehoods other countries are far more involved. The world is a big place with some legal jurisdictions more rigorous than others. Not all good but not all bad either.
Powerful anonymous libelous contributors that were producing heavily biased content would see the foundation itself being targeted for legal action. It might be slow in happening due to cognitave momentum and vanity but it happens. One thing that is an issue is that Wiki tends to play safe legally a bit too often.
Yeah, it's a real problem with Wikipedia..
I captured the fact friend; every on the internet is looking for information not website'. AI has closed this gap and unless Wikipedia can evolve, the situation will be worse by next year .
Very relevant and thought-provoking article! It clearly shows how AI is changing Wikipedia and the way we search for information.
Very relevant and thought-provoking article! It clearly shows how AI is changing Wikipedia and the way we search for information.
Well nothing ultimately lasts forever. I think some of these companies can already see the handwriting on the wall quite clearly
The shift is understandable. Why do we need traditional models if AI can deliver instant and contextualized answers?
!BBH
!PIZZA
!LOLZ
This is false. Editors update information based on reliable sources. When there is a dispute then the editors work on consensus to determine what is likely the most accurate information source.
Wikipedia is under attack by those who want to push 'alternate facts' to suit their own ideology.
For anyone who isn't sure... you can check Wikipedia yourself. I found that Fox News was used as a reference 18K times, which was more than MSNBC and Mother Jones. They try exceptionally hard to be unbiased, but there is a smear campaign against it by conservatives because they want anyone to have access to reliable information sources.
If AI successfully eats Wikipedia, then AI will no longer have a reliable source of information and will hallucinate even more than it already does.
I imagine Grokipedia will be wildly unreliable.
$PIZZA slices delivered:
@rzc24-nftbbg(1/10) tipped @taskmaster4450
Come get MOONed!
Man just shows you where onto something with trying to build a wiki on Hive.
Great news. Wiki is overwrought with ideology.