5 Minutes
Something strange has been happening to the internet’s traffic lights. The links are still there. The articles are still being published. Yet millions of readers who once clicked through to tech news sites simply… stop at Google.
The culprit appears to be Google’s AI Overviews — those tidy, AI‑generated summaries that now sit at the top of many search results. Instead of sending curious readers to the original reporting, the system digests dozens of articles, condenses the information, and delivers a quick answer directly on the results page. Convenient for users. Devastating for publishers.
For years, search traffic acted as the economic backbone of digital media. A reader searches a question, clicks a link, lands on a publication’s site, and advertising or subscriptions help fund the reporting. AI Overviews quietly disrupt that loop by answering the question before the click ever happens.
When answers replace visits
New analysis from SEO firm Growtika suggests the shift is already dramatic. Using Ahrefs data, the company tracked Google search traffic to ten major technology publications between early 2024 and early 2026.
At their peak, those outlets collectively received roughly 112 million monthly visits from U.S. Google searches. By January 2026, that figure had collapsed to just under 50 million.
The losses were not evenly distributed. Mashable experienced the mildest decline, though “mild” is relative — the site still lost about 30 percent of its search traffic. Others were hit much harder. Wired reportedly saw a 62 percent drop. Publications such as How‑To Geek, The Verge, and ZDNet each lost more than 85 percent of their Google‑driven visits during the two‑year period.
One case stands out as particularly stark. Digital Trends, once pulling in around 8.5 million monthly clicks from Google in March 2024, reportedly dropped to just 264,861 by January 2026. That represents a staggering 97 percent decline in U.S. search traffic.
Put another way: according to Growtika’s comparison, the four hardest‑hit tech sites now attract less combined monthly traffic than the single Reddit community r/ChatGPT.
Growtika is careful not to claim a single cause. Traffic patterns on the internet are notoriously complex, and multiple forces are reshaping how people discover information online.
Still, the firm points to three overlapping shifts. First, the rollout of AI Overviews itself, which began appearing broadly in mid‑2024. Second, changes to Google’s ranking system that increasingly surface Reddit discussions and forum posts near the top of results. And third, the rapid adoption of standalone AI chatbots, where users search for answers without touching Google at all.
The timing lines up with another key moment. In mid‑2025, Google expanded AI Overviews to cover a far wider range of search queries. By July of that year, the feature appeared in roughly one quarter of all searches — dramatically increasing the number of times users received an instant AI summary instead of a list of links.
Google disputes the conclusions. In response to the analysis, a company spokesperson described the methodology as “fundamentally flawed,” arguing that the dataset examined too few websites and failed to account for normal seasonal fluctuations in traffic.
The company also notes that audiences are increasingly consuming information in new formats, from podcasts to community forums, rather than traditional articles.
Yet the broader question remains uncomfortable for the media industry. If search engines summarize journalism instead of sending readers to it, the economic model that funded much of the open web begins to wobble.
AI tools may make information faster to access, but they also risk severing the link between the people who produce knowledge and the audiences who rely on it.
And if that link breaks, the internet could slowly transform from a web of sources into a feed of answers — efficient, polished, and increasingly detached from the original reporting that made those answers possible.
Source: futurism
Leave a Comment