🔍 Read the full analysis: What A Data Check Reveals About OpenAI’s ChatGPT Search Claim on ThorstenMeyerAI.com
Get hardware and tech essentials delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
OpenAI executive Adam Fry said ChatGPT’s search-like query volume is roughly one-fifth of Google’s, but the company did not disclose the Google figure, time period, countries or underlying data. Independent estimates are lower, while website traffic panels show Google sends far more visits than ChatGPT. Those measures track different activities and do not independently verify OpenAI’s claim.
OpenAI consumer product head Adam Fry said ChatGPT handles roughly one-fifth of Google’s search-like query volume, but he did not identify the Google benchmark or provide data that would let outsiders verify the comparison. A review of published usage figures, third-party estimates and website traffic panels finds that the numbers describe different activities: estimates of search-like use are lower, and Google sends substantially more visitors to websites.
Fry made the statement during a fireside chat with journalist Alex Kantrowitz at an OpenAI-hosted dinner in New York on November 2. Big Technology published a transcript the following day. According to the source account, Fry defined a search as each time someone asks a question and ChatGPT searches the web. The Verge reported that a definition or simple how-to answer without a web lookup would not count. Fry also said ChatGPT had grown from one billion to 1.2 billion weekly users in two months, but did not supply supporting figures in the reported exchange.
The comparison leaves key details unspecified: which Google query total OpenAI used, the measurement period, the countries included and the data behind ChatGPT’s count. The Verge reported that Fry did not specify the source of Google’s search-volume figure. Google did not comment on the claim in the reports cited by the source material. The remark is confirmed as something Fry said; the ratio itself is not independently confirmed.
Third-party estimates cited in the review put ChatGPT’s search-like activity below the one-fifth figure, but each relies on modeling rather than a direct count. Ahrefs estimated about 12 percent in February 2026, using its judgment about which prompts resembled searches. seoClarity estimated about 7.5 percent on October 28, using projected ChatGPT message volumes and assumptions about which categories count as search. Neither estimate is directly comparable to OpenAI’s undisclosed calculation.
Infographic 01 · ChatGPT as a share of Google
Questions or clicks? ChatGPT is a fifth of Google, or less than 2 percent
OpenAI says one fifth. Independent models say less. Traffic panels say far less, because a chat answer often ends without a visit.
Counting questionsHow many search-like questions?
Counting clicksHow many visits reach websites?
All values: ChatGPT as a percentage of Google on the same yardstick. OpenAI’s figure is a company claim (striped bar); the two models are estimates; the panels see only their own sites. Bars scaled to 20 %.
Sources: Big Technology, 3 Nov 2026 · Ahrefs, 11 Feb 2026 · seoClarity, 28 Oct 2026 · Ahrefs traffic panel, Oct 2026 · StatCounter, Oct 2025–Oct 2026.
Powered by Thorsten Meyer AI
Search Use Does Not Equal Website Reach
For publishers, retailers and other businesses, the distinction between a question asked and a visit delivered matters. Search-like use can show that people ask ChatGPT for information, but it does not establish how often ChatGPT sends those people to the original websites. In one October panel covering 118,671 websites, Ahrefs recorded Google at 24.74 percent of all visits and ChatGPT at 0.36 percent. That is a measure of traffic in that panel, not a count of searches across the internet.
A separate StatCounter comparison measured clicks from search engines and chatbots to sites in its network. From October 2025 through October 2026, ChatGPT’s share of the Google-versus-ChatGPT clicks ranged from 0.08 to 0.41 percent per month. StatCounter’s figures concern referrals, not the number of questions people ask. They cannot disprove an estimate of search-like queries, but they show why a headline ratio alone does not answer how much web traffic AI search generates.
The distinction also affects how companies interpret changing discovery habits. ChatGPT citations may matter to whether a source is seen, while referral figures indicate whether users visit it. The available measurements support tracking both kinds of exposure, but they do not establish that AI answers have replaced conventional search traffic. Users, meanwhile, still need to check cited sources when accuracy or the original context matters.
Products Worth Considering
Bose QuietComfort Wireless Noise Cancelling Headphones, Black
Sony WH-1000XM4C (M4 2nd Gen) Wireless Noise-Cancelling Headphones, Black
Bose QuietComfort Headphones (2nd Gen), Wireless Over-Ear Noise-Cancelling Headphones, Black
As an affiliate, we earn on qualifying purchases.
Different Measures, Different Ratios
The figures in the review use at least two different definitions. OpenAI’s claim concerns search-like queries that trigger a web lookup, though its comparison method is not public. Ahrefs and seoClarity estimate how much ChatGPT activity resembles searching, each applying its own assumptions. The traffic panels instead count visits or clicks reaching websites. Those measures answer related but separate questions, so their ratios should not be treated as competing readings of one common statistic.
OpenAI has also published consumer usage data in which “seeking information” accounted for 20.2 percent of messages in September 2026. That category is not equivalent to web searches: seeking information can happen without a web lookup, and other categories might include messages that do trigger one. The source review argues that its own rough arithmetic would require about six in ten ChatGPT messages to trigger web search for the one-fifth comparison to fit, but that is an inference, not a disclosed OpenAI calculation. The public category data therefore raises a question without settling it.
Google’s own reported performance offers additional context, not a direct check on Fry’s ratio. Alphabet said Google query growth in the third quarter was its fastest in five years and reported search revenue of $63.35 billion, up 12 percent. Revenue growth and query growth do not reveal the exact search volume OpenAI used. They do show that the comparison is being made against a search business that remains large and was still growing in the period described.
“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, as quoted in Big Technology
The Missing Benchmark and Method
OpenAI has not publicly specified the Google query total or measurement window behind the one-fifth comparison, nor the countries covered or how it counted ChatGPT web lookups. Without those details, outside researchers cannot reproduce the calculation or determine how it compares with the estimates in the review. The source material does not establish whether OpenAI has supplied this information elsewhere.
The independent figures have limits too. Ahrefs and seoClarity model search-like activity using different assumptions, while Ahrefs and StatCounter traffic panels cover their own website samples and record referrals rather than total queries. OpenAI’s published “seeking information” category is not a count of web searches. As a result, the available data cannot produce a single apples-to-apples figure. The one-fifth ratio remains an attributed company claim, not a verified market-share statistic.
What Data Could Settle the Comparison
A more testable comparison would require OpenAI to publish its query definition, time period, geographic scope and underlying counts, along with the specific Google figure used as the denominator. Google’s public comment or a comparable independent measurement could help clarify the basis, but no such confirmation is included in the reports cited here. Until fuller methodology is available, third-party estimates should remain labeled as models and traffic-panel results as referral measurements.
For now, the reported figures point to separate measures rather than a settled contest over search volume. Businesses can monitor AI citations and referrals alongside conventional search traffic, while readers can follow cited links when they need to assess a source directly. Whether ChatGPT’s query volume is actually one-fifth of Google’s—and how much of that use sends people to the open web—remains unresolved.
Key Questions
What did OpenAI claim about ChatGPT and Google?
Adam Fry said ChatGPT’s search-like query volume was roughly one-fifth of Google’s. He did not identify the Google figure or publish the calculation in the reported remarks.
Has the one-fifth figure been independently verified?
No. The statement is confirmed as Fry’s claim, but the underlying ratio is not independently verified. OpenAI did not provide enough methodological detail in the reported exchange for outsiders to reproduce it.
Why do the estimates differ?
They use different definitions and methods. Some estimate which ChatGPT prompts resemble searches; others count website visits or clicks. Those are not interchangeable measures, and modeled estimates also depend on their authors’ assumptions.
Does the traffic data show how often people search with ChatGPT?
No. The cited panels measure referrals to websites in their networks, not every ChatGPT question or web lookup. They indicate that Google sent much more measured website traffic in those panels, but do not settle the total query-volume comparison.
What information is still missing?
OpenAI has not specified the Google benchmark, time period, countries or underlying counts used for the ratio in the reported remarks. A clear methodology and comparable data would be needed to assess the claim directly.
Source: ThorstenMeyerAI.com
