News Sites Are Dying To AI Search Engines, And They Don’t Care!

If you’re reading this, then congratulations! You’re one of the diminishing number of people who actually gives a damn about news websites. We’re at the beginning of a new age of the internet, one where news websites are no longer your primary source of information. AI search engines and AI tools are sucking the life out of news websites and regurgitating it onto the web via AI-generated responses. What’s worse is that the companies know that they’re killing these websites, but they don’t care about it at all.

We’re all getting sick and tired of seeing CEOs of companies like Google, Microsoft, OpenAI, and others standing on stage and pretending that all is right in the world of AI. They smile and talk about all of the “benefits” that AI is bringing. However, they blatantly ignore all of the downsides. One of the downsides is how these AI search engines are screwing over the thousands of news sites that rely on traffic.

How deep is the cut?

A new report shines a rather harsh light on search engines like Perplexity and ChatGPT Search and AI tools like AI Overviews. A company named Tollbit analyzed over 160 websites between October and December 2024. Forbes reported on this news.

These tools send 96% less traffic to websites than traditional search engines. Let that sink in. This means that companies that rely on search engines to send traffic their way are seeing devastating drops in their click-through. Companies are being gutted because they don’t get enough traffic to stay afloat.

And what do these companies do in response? The same thing they always do. They either blatantly lie to us (like Google, which constantly talks about how AI Overviews sends viewers to more sites whenever a company sues it) or they favor big brands as a pathetic attempt to seem fair to publishers (like Perplexity and its revenue-sharing program). The worst part is that neither company wants to actually acknowledge that sites are suffering. All that matters is that they attract the attention of lazy people for whom doing a quick web search is suddenly too much work.

None of the companies want to walk back their AI tools because they have too much to gain. If 90% of the internet-based companies shutter, it wouldn’t matter because their AI can still generate whatever people need. So, it’s all good…

Casualties

Chegg, an education company, recently made headlines for suing Google for its AI Overviews. The company saw a massive 49% drop in traffic in January this year, and this was after suffering an 8% drop in Q4 2024.

However, Google stated that “AI Overviews send traffic to a greater diversity of sites,” in response to the lawsuit. Really, Google? Websites across the board are seeing drops in traffic, how is that sending traffic to a “greater diversity of sites“?

This problem doesn’t only exist today. Back in 2023, when Google was still testing AI Overviews under the name “Search Generative Experience”, Jake Boly, who writes show reviews, saw a 96% drop in his traffic. It doesn’t take a mathematician to know that that’s not sustainable.

The age-old issue persists

As we speak, internet-based publications are disabling their websites and getting ready to shut down without so much as a “thank you” from the companies that stole their data. These companies are still scraping data from around the internet without giving the sites anything in return. They’re stealing the data that they produce and using it to train the very tools that are putting them out of business. That’s the biggest slap to the face.

Only the biggest companies can bring these AI companies to court, like The New York Times or Forbes, but most of the smaller companies can’t do anything to fight it. We’d like to see Sundar Pichai address this during Google I/O. However, we know that he’s going to talk about how many millions of people use Gemini to “improve” their lives.

The web scraping is costing websites even more

Not only are companies not giving websites any sort of compensation for scraping their data, but they’re actually costing these companies money. Big AI companies like Google, Meta, OpenAI, Perplexity, and others need to scrape data in order to deliver real-time data to their search engines. That data comes from the websites they’re stabbing in the back.

Between October and December last year (Q4 2024), major tech companies like Meta, OpenAI, and Perplexity scraped websites upwards of 2 million times, as per the Tollbit report. Several of the pages received up to seven visits. So, much of the traffic that these sites are getting are from bots that don’t contribute to the ad revenue.

That’s not the worst part, as these visits drive up server costs for the sites. So, these sites end up paying more in the end, along with having their data and traffic stolen from them. TollBit CEO Toshit Panigrahi stated, “AI does not read like humans do. Humans will click one link, they’ll click the second link and then they’ll move on…AI will read 10 to 20 links to get their answer.” Again, we’d like to see the lengthy blog post from Sam Altman mention that rather than how many hours until we reach AGI.

AI companies are killing this industry, but they just don’t care

Hundreds of sites are dying, and major tech companies are holding the smoking guns. Yet, they deny pulling the triggers. The worst part is their blatant disregard. They see that their actions are hurting the publishing industry, yet they double down on the technology that’s doing the damage.

When Google unveiled AI Overviews, it was widely panned. It responded by pushing it to the masses and it claimed it did no wrong. Meta stopped linking to news websites and had Meta AI generate search results. Perplexity is making deals to share revenue with a short list of big brands and leaves little sites in the dust. It almost feels like these sites banded together specifically to put the publishing industry in the ground.

There’s no telling when we’ll start to see a palpable decrease in news coverage thanks to these practices. Sure, OpenAI is partnering with news publishers, but how long can it pay these companies millions to scrape their data? Sooner or later, the dozens of websites that sold out to OpenAI will start to feel the impact. At a certain point, the news industry will be dead, and all we’ll have will be quick AI summaries with varying levels of accuracy. This isn’t the kind of future that companies talk about, but it’s the future that we’re headed toward.