People Aren’t Leaving Google For AI. They’re Using Both
Similarweb finds 95% of ChatGPT users still use Google. Other research finds fewer searches and fewer clicks. Here’s how both can be true at the same time.

Most ChatGPT users also visit Google. Similarweb data shows that 95% of ChatGPT users were also part of Google’s audience in September 2025, a share that held steady through May 2026 even as visits to generative AI platforms grew 70% year over year. Similarweb interprets this as AI being layered on top of search rather than replacing it. Greg Jarboe highlighted this in his July SEJ review.
A study from Bocconi University, covered here a month earlier, found a 9.4% drop in traditional search queries among households that gained access to ChatGPT Search, with the decline growing the longer they had access. Both findings can be true because overlap counts people, whereas the Bocconi study counts queries per household.
This year’s research consistently shows those two measures moving independently, along with a third—clicks—which can fall even when the number of people and queries holds steady. A search practitioner trying to explain why referrals dropped while reports say nobody left Google is standing right in that gap.
Using Google Is A Low Bar To Clear
Overlap measures whether the same individuals are part of two different audiences during a specific time frame. For example, someone who performs a single Google search in a month is categorized as a Google user. This could include someone who has shifted some research to ChatGPT but still relies on Google for maps, login, and store hours.
Similarweb’s August post on the report clarifies that overlap looks at whether the same people appear in both audiences, not how they interact within a single research session. Someone can remain a Google user while sending a growing share of specific query types to ChatGPT. The report doesn’t specify query-level substitution, a point confirmed in the post. It also excludes API usage, desktop apps, and in-app AI features, which Similarweb says are more likely to replace search activities than add to them. A developer using the API or someone doing all their work through a desktop app wouldn’t appear in the panel.
Overlap doesn’t indicate frequency of use. Pew’s February survey of 5,119 U.S. adults found that 49% have ever used a chatbot, 42% use one to find information, and 60% read AI summaries at the top of search results. These statistics were covered in June. “Ever use” is a broad metric, and these figures don’t detail how usage is distributed across different activities.
Same Users, Fewer Searches
The Bocconi paper examines how search behavior shifts when households gain access to ChatGPT Search. Using Comscore’s U.S. desktop clickstream data from October 2024 to July 2025—a window covering OpenAI’s three expansions of ChatGPT Search access—the study compares households that gained access with similar households that had no prior ChatGPT or Claude activity.
Results show that, on average, households reduced their traditional search queries by 3.14 per week after gaining access, roughly 9.4% less than the pre-expansion average of 33.51. The decline deepened over time, reaching 17.0% after 20 weeks.
Comparing logged-in ChatGPT users with Search access in December 2024 to anonymous users who gained access in February 2025 showed smaller declines: 4.9% on average and 8.2% after 20 weeks. The authors interpret the smaller figure as evidence that comparing adopters with nonusers may capture differences in who adopts AI rather than the effect of access itself.
The largest drops in search referrals came in informational categories. Referrals to academic sites fell 32.8%, reference sites 26.5%, developer sites 15.1%, and news sites 13.4%. In contrast, referrals to marketplaces and entertainment sites barely moved or were indistinguishable from zero. The paper stresses that these patterns are descriptive and don’t establish causality.
According to the same Comscore panel, Google’s household reach held steady at about 49%, while ChatGPT’s grew from 4.6% to 7.9%. Households with access still ran fewer queries, so in one dataset Google’s audience held while queries fell. These insights come from a working paper analyzing desktop browsing data up to July 2025, before AI Mode reached its current scale.
Other Research Finds Search And AI Used Together
A working paper by Samira Gholami, Cristiana Firullo, Cristobal Cheyre, and Alessandro Acquisti, posted on SSRN in February, reports that users who adopted a chatbot visited more websites afterward. They studied 4,394 Comscore desktop users who started using a large language model (LLM) between August 2023 and January 2024. Adoption was linked to a consistent increase in weekly unique website visits and a temporary boost in traditional search activity that later declined.
The authors acknowledge the search estimates are noisy and stop short of claiming a lasting reduction. Data from a 2025 field study with 308 U.S. Chrome users points in the same direction: sessions using both an LLM and search engine lasted longer and involved more sites, while LLM-only sessions were shorter and narrower. Since the Comscore data predates ChatGPT Search and AI Mode, it concerns earlier products than those discussed by Bocconi.
The team behind the AI Mode experiment reviews these earlier studies in their own paper and describes the evidence as mixed: some studies show people searching less after adopting a chatbot, while others indicate they use both side by side.
AI Mode Can Replace The Results Page Without Replacing Google
Google’s own numbers make it harder to treat Google and AI as separate entities. At I/O in May, Sundar Pichai announced that AI Overviews had over 2.5 billion monthly active users and AI Mode surpassed 1 billion in its first year. Liz Reid also noted record query volumes last quarter. Pichai summarized the company’s view: “When people use our AI-powered features in Search, they use Search more.”
These claims lack baseline or user-level data for independent verification. What the numbers do show is that a growing share of Google usage runs through AI features, covered when the first AI Mode usage data came out. When users engage with AI Mode, they are both Google and AI users, and may never see the traditional results page.
A field experiment, published on arXiv on August 18, examined what happens when AI Mode becomes the primary experience. Stephanie T. Wang, Jeffrey Gleason, Yakov Bart, Christo Wilson, and Danaé Metaxa recruited 1,100 U.S. Chrome users who rely on Google as their main search engine. After three days of observation, they randomly assigned participants to one of three groups for a week: standard Google, Google with AI Overviews hidden, or Google with all searches routed through AI Mode.
At baseline, 0.6% of searches used AI Mode, but in the treatment group, 94.7% of searches were routed there—what the authors describe as forced, full adoption rather than the gradual opt-in most people would experience. They also noted that the effects of voluntary use may differ.
In this setup, click-through to external sites fell by 18.8 percentage points (95% confidence interval: -22.2 to -15.3), and search sessions declined by roughly 0.92 per day from an average of 4.0. More frequent Google users experienced larger declines. Average session duration increased by 0.43 minutes. The share of participants clicking through to news sites dropped by 12.5 points, Reddit by 21.2 points, and Wikipedia by 9.9 points.
Participants also explored alternatives. The share using Bing, DuckDuckGo, or Yahoo rose by 11.2 points, but ratings for trust, satisfaction, and usefulness fell. Of 309 AI Mode users who answered an open-ended question, 47 (about 15.3%) reported difficulty reaching specific websites, with one person noting that when they entered a site, “it would tell me about the website instead.” When AI Overviews were hidden, external clicks increased by 8.8 points among participants where the extension worked. A mid-study change to Google’s interface disrupted the hiding feature for roughly half of the AI Overviews group.
The study sample skewed younger, more educated, and more left-leaning than the general U.S. population. A seven-day period is also too short to capture long-term habit changes. In April, an experiment by Agarwal and Sen also routed a group into AI Mode and found fewer outbound clicks and lower satisfaction, though those authors labeled that arm exploratory due to participant dropout. Wang’s team pre-registered their AI Mode comparison and found no evidence of increased dropout, so the two experiments now point in the same direction.
Why This Matters
A news website can maintain consistent Google impressions while referral traffic declines, consistent with research findings. Bocconi reported a 13.4% drop in search referrals to news sites among households with ChatGPT Search access, while Wang observed a 12.5-point reduction in users clicking news links under AI Mode assignment. In Bocconi’s study, Google’s audience held steady despite fewer searches; in Wang’s data, users kept searching on Google but clicked fewer links. The pattern differed for marketplace and entertainment sites in Bocconi’s results. Referrals to those sites stayed nearly flat after ChatGPT Search access, while referrals to reference and academic sites dropped by more than 25%. The paper stops at describing that split. Wang’s study also found some people struggled to reach specific websites of any kind under AI Mode. When all traffic is aggregated, these differences become less visible.
Agencies frequently hear the 95% figure from clients, usually based on headlines. Similarweb sees the continued overlap as evidence that people haven’t abandoned Google entirely, at least within the audience it monitors. But that number doesn’t reveal how many queries those users are making. The closest thing to a controlled test of that question found traditional search queries declined on U.S. desktops after ChatGPT Search access, particularly for informational queries.
When someone asks ChatGPT for a recommendation and then searches for that brand on Google, it appears as an organic visit in Search Console. However, there’s no straightforward way to determine the actual source of that visit. That’s the in-house team’s problem, and the overlap number hides it. Similarweb names branded search lift as a metric that would move if AI were feeding demand, calling it a type of conversion most analytics tools don’t typically attribute to an AI session.
Metrics like audience size, query volume, search sessions, AI visibility, and referral traffic are often discussed as if they’re part of the same measurement. This year’s data shows that while Google keeps users engaged, it’s losing some queries—and it can keep the query while losing the click. In Wang’s study, people kept searching while clicks and trust fell.
What None Of These Datasets Can See
None of the datasets discussed track individual users across platforms like Google Search, AI Mode, ChatGPT, and Gemini over a long period at the task level. Similarweb focuses primarily on audience data rather than detailed session information. Bocconi and Gholami analyze U.S. desktop clickstream data, with Bocconi’s dataset ending in July 2025. Wang’s experiment follows roughly a thousand Chrome users for seven days, during which their searches were routed into AI Mode.
Google has not released user-level data to verify its claims. Its I/O posts highlight record-high query volumes and increased search activity but don’t specify what constitutes a query in AI Mode, provide a baseline, or clarify which queries increased.
It’s also unclear which types of searches are moving to AI versus staying in a search box. Bocconi’s split between informational and transactional categories is descriptive only. Wang’s team notes their experiment wasn’t designed to detect small effects in secondary click outcomes, and Gholami’s session categories derive from data gathered from 308 users.
Looking Ahead
During Wang’s March experiment, AI Mode didn’t show ads, but the authors expect satisfaction and click metrics to shift once ads are introduced. On May 20, Google announced it is testing two new ad formats built with Gemini in AI Mode and expanding its Direct Offers pilot to include AI Mode responses. The earlier data on clicks and trust comes from a previous product version. All three academic papers remain drafts and haven’t been peer-reviewed yet. Updates will follow if findings change in review. Alphabet’s upcoming earnings call, typically in late October, is where Google is likely to revisit its query-growth claim—and the key detail to listen for is whether it comes with a baseline this time.
More Resources:
- Google CEO Sundar Pichai Is OK With AI Mode Replacing Classic Search
- Google Data Compares Gemini & AI Mode Use Against Daily Life
- 846,000 Google Searches Reveal How AI Overviews Are Changing User Behavior
Featured Image: Tada Images/Shutterstock

