TL;DR
Prime made for students and young adults
- Fast, free delivery for dorm and study essentials
- Prime Video and Amazon Music included
- Member-only deals
OpenAI consumer product head Adam Fry said ChatGPT handles roughly one-fifth of Google’s query volume, counting questions that trigger a web search. OpenAI has not published the Google baseline or data behind the comparison. Independent estimates put the share lower, while website traffic panels show Google sending far more visits than ChatGPT.
OpenAI consumer product head Adam Fry said November 2 that ChatGPT handles roughly one-fifth of Google’s query volume when counting questions that prompt a web search. The figure is a company claim: OpenAI has not identified the Google baseline or released underlying data, and independent estimates and website traffic panels measure different things and show smaller shares.
Fry made the comparison during a fireside chat with journalist Alex Kantrowitz at an OpenAI-hosted dinner in New York. Big Technology published the conversation on November 3. The Verge, which described the gathering as a press event, reported that Fry did not specify the source of the Google search figure. Google did not comment on the claim in the reports cited by the source material.
Fry’s definition was narrower than counting every ChatGPT message. He said, “Every time you ask a question and go search the web, that’s a search.” The Verge reported that requests such as defining a word or giving a simple how-to explanation do not count under that definition. Fry also said ChatGPT had grown from one billion to 1.2 billion weekly users in two months; the material provided does not include independently verified support for that figure.
Two independent estimates cited in the source material are below OpenAI’s ratio. Ahrefs estimated ChatGPT’s search-like prompts at about 12% of Google’s volume in February 2026. SeoClarity estimated about 7.5% on October 28. Both are modeled estimates, not direct counts of all searches. Their methods and definitions differ from the measure Fry described, so the figures are not direct tests of the same quantity.
Website referrals provide another, distinct measure. Ahrefs said its panel of 118,671 websites received about 69 times as much traffic from Google as from ChatGPT in October 2026. That implies ChatGPT supplied roughly 1.5% as much traffic as Google within that panel. StatCounter’s monthly figures for October 2025 through October 2026 put ChatGPT’s share of clicks to sites in its network between 0.08% and 0.41% of Google’s, according to the supplied material. These panels capture visits to their observed sites, not every question asked or every search conducted.
Does ChatGPT handle one-fifth of Google searches? The data has gaps
OpenAI made the comparison, but has not published the baseline or data behind it. Independent estimates and website traffic panels measure different things.
Search Questions and Website Visits
The figures matter because query volume and referral traffic answer different questions. ChatGPT can answer a user without sending that person to a website. A count of questions that trigger web lookup may therefore be much larger than the number of visits publishers receive from ChatGPT. The one-fifth comparison, even if accurate within OpenAI’s definition, does not establish that ChatGPT is delivering one-fifth of Google’s website traffic.
For publishers and businesses, the available evidence points to a need to track both search visibility and AI referrals. A site may be mentioned or used as a source in an AI answer without receiving a click. The supplied material does not quantify how often this happens or what it means for sales, readership, or other outcomes, so those effects remain unmeasured here.
For users, the distinction also affects how to judge answers. A response that includes sources still depends on the material behind it. When a claim matters, opening a cited source can help readers check its wording and context. The figures presented here do not measure the accuracy of ChatGPT’s answers.
Popular right now
De'Longhi Dinamica Plus Fully Automatic Espresso Machine
Dyson V8 Cyclone Cordless Vacuum
Samsung QN65LS03FAFXZA 65 The Frame QLED 4K Art Mode Vision AI Smart TV (2025)
As an affiliate, we earn on qualifying purchases.
What Each Search Measure Counts
The comparison came amid broader discussion about whether AI assistants are taking activity from search engines. But “search volume” has no single measure across the figures cited here. Fry described ChatGPT questions that lead to a web lookup. Ahrefs and seoClarity modeled prompt categories. Ahrefs’ traffic panel and StatCounter counted referrals or clicks observed across their networks. Each measure captures a different part of user behavior.
The source material says Ahrefs’ February estimate used prompts of a kind people might otherwise use on Google. It describes seoClarity’s estimate as combining information-seeking prompts with a portion of practical-guidance prompts. The available excerpt does not provide the complete calculations or limitations for either model. Their percentages should be read as estimates tied to those methods, not as audited totals of ChatGPT and Google searches.
Google remains a large source of traffic in the cited panels, and its own business figures provide separate context. Alphabet said Google query growth in the third quarter of 2026 was its fastest in five years, while search revenue rose 12% to $63.35 billion. Those company-reported figures do not establish how ChatGPT compares with Google in query volume, but they show that Google’s search business was still growing over the reported period.
““ChatGPT’s volume of search-like queries is now roughly one-fifth the volume of queries of Google.””
— Adam Fry, head of ChatGPT’s consumer product, quoted by Big Technology
The Missing Baseline and Methods
The central gap is that OpenAI has not named its Google denominator. The company did not specify the period, countries, or underlying data behind the one-fifth figure in the reports cited. Without those details, outside analysts cannot reproduce the calculation or determine how closely its definition matches Google’s own query counts.
The other percentages do not settle the question because they use different methods. The two search-volume estimates are models, while the traffic panels count activity observed on specific sets of websites. The source material does not provide full methodological detail for each estimate, nor does it establish that any one panel represents the entire web. The figures support a distinction between estimated query activity and observed referrals, rather than a single definitive ratio.
It also remains unclear whether Fry’s reported user growth and search-like query ratio cover the same period or user base. The supplied material records those statements but does not include a data release that would let readers check them independently. Google’s response to the comparison was not available in the cited reports.
What a Verifiable Comparison Needs
A clearer comparison would require OpenAI to publish the time period, geographic coverage, query definition, and calculation behind its estimate, along with the Google figure used as the baseline. Comparable definitions from both companies would make it easier for outside researchers to assess whether the ratio describes web searches, questions that trigger a lookup, or another measure.
Further reporting can also track modeled search activity and observed website referrals separately as new data becomes available. The figures cited here cover different periods, including estimates from February and October 2026 and traffic observations for October 2026. They do not establish how the relationship will change in later months. For now, the one-fifth ratio remains OpenAI’s stated comparison, while independent measures show lower estimates or far fewer visits to the sites those panels track.
Key Questions
What did OpenAI say about ChatGPT and Google searches?
Adam Fry said ChatGPT’s volume of search-like queries was roughly one-fifth of Google’s query volume. He defined a counted search as a question that leads ChatGPT to search the web.
Has the one-fifth figure been independently verified?
No. OpenAI has not disclosed the Google baseline, period, geographic coverage, or underlying data cited in the reports. The statement is confirmed as something Fry said, but the ratio itself has not been independently checked in the supplied material.
What do the independent estimates say?
Ahrefs estimated about 12% of Google’s volume in February 2026, and seoClarity estimated about 7.5% on October 28. Both are models with their own definitions, not direct counts that reproduce OpenAI’s method.
Does one-fifth of Google’s query volume mean one-fifth of its website traffic?
No. Queries and visits to websites are different measures. Ahrefs’ October 2026 panel reported about 69 times more traffic from Google than from ChatGPT across 118,671 websites. That panel measures its observed referrals, not all queries.
What information is still missing?
OpenAI has not identified the Google number used for comparison or specified the period, countries, and underlying calculation. The source material also does not provide enough methodological detail to make all the independent estimates directly comparable.
Source: Thorsten Meyer AI
