OpenAI says ChatGPT search volume is one-fifth of Google’s | Magica
OpenAI puts ChatGPT’s search-like queries at one-fifth of Google’s
Editorial Team
••📖7 min read
OpenAI says ChatGPT handles search-like queries equal to roughly 20% of Google’s volume. The claim signals a growing competitor for information queries, but undisclosed counting methods prevent a clean market-share comparison.
OpenAI says ChatGPT’s search-like query volume is roughly one-fifth of Google’s.
The company has not supplied enough methodology to turn that ratio into market share.
Growing usage leaves separate questions about displacement, advertising revenue and computing costs.
OpenAI says ChatGPT, its AI assistant, now handles search-like queries equal to roughly 20% of Google’s query volume. That would make it a substantial destination for finding information, but the disclosure does not establish how much business it has taken from Google’s search engine.
Adam Fry, OpenAI’s consumer product lead for ChatGPT and a former Google employee, made the comparison in an interview published November 3, conducted at an OpenAI-hosted dinner in New York. OpenAI also credits Fry among the team behind ChatGPT search’s original launch. He said ChatGPT had reached 1.2 billion weekly active users, up from 1 billion in two months. Those are company-provided figures, rather than independently verified measurements.
The stakes extend beyond which service answers a question. OpenAI is building subscriptions, advertising and paid credits around an assistant that can retain conversational context and undertake tasks. Search-like usage gives it more opportunities to serve those customers, but query counts alone say little about the revenue or cost of doing so.
OpenAI’s October 2024 launch demonstration shows ChatGPT search with linked references and a citations sidebar. Source: OpenAI. Source: OpenAI.
What the one-fifth comparison measures
Fry described search-like queries as requests people would previously have taken to a search engine, including questions needing current information. Pressed on whether the unit was a conversation or each question requiring web access, he said:
Every time you ask a question and go search the web, that’s a search.
That definition is narrower than counting every ChatGPT message. It also allows a conversation containing multiple web-search questions to contribute multiple searches. It does not explain how those requests are made comparable with Google queries.
The account of the disclosure notes that OpenAI did not reveal the source of its Google volume figure. The interview supplies neither absolute search totals nor a comparison period or geographic scope. Fry also did not substantiate the interviewer’s suggested comparison with Bing. A ratio to Google’s volume is different from a share of all searches across competing services, and does not establish a ranking against them.
A national exit poll found broad opposition to new local data centers and a Democratic tilt among voters who prioritized AI. But the issue ranked below four other tested concerns, industry-backed candidates also won, and the next fight is over siting and power costs.
Editorial Team
There is a public baseline, but it cannot repair those gaps. Google said in a May 2025 announcement that it handled more than 5 trillion searches annually, citing internal data from January 2025. That earlier figure is not an identified denominator for OpenAI’s November 2026 claim; using it to calculate a current daily ChatGPT total would imply comparability the disclosure does not establish.
OpenAI’s launch record distinguishes the original release from later access expansions. Graphic: Magica; source: OpenAI. Source: OpenAI.
Search grew out of an existing assistant
ChatGPT initially drew users for writing and coding help, Fry said; requests requiring current information broadened its uses. OpenAI’s launch announcement dates ChatGPT search to October 31, 2024. It incorporated features from SearchGPT, an earlier prototype, into the assistant.
Fry described the search effort as having begun about a year and a half earlier. The initial launch was roughly two years before this interview, with access expanding to all logged-in users in December 2024 and to everyone in supported regions, without signup, in February 2025. Those are distinct milestones.
From launch, ChatGPT could choose to search based on a question, or users could select web search themselves. OpenAI said it used third-party search providers and content supplied directly by partners. The usage claim measures activity inside ChatGPT; it does not establish an independently built search index or describe the current infrastructure behind every request.
OpenAI’s record with separate products is less uniform. A report examining the claims noted that ChatGPT Atlas, its standalone web browser, had been retired earlier this year. Growth inside the assistant does not establish success for every attempt to compete with Google’s products.
More ChatGPT usage need not mean less Google usage
Fry argues that remembering the broader context of a conversation makes ChatGPT useful for information requests. A question about a landmark, for example, can sit within an ongoing discussion of a trip rather than stand alone. He also reported 250 million weekly active users of voice input, including dictation and spoken conversations. That figure measures another way of using ChatGPT; it is not a count of search users.
Google already offers a conversational alternative inside its own search business. In the same May 2025 announcement, it described AI Mode as supporting follow-up questions and links to the web, and said it was beginning to test ads within that experience. Google also said AI Overviews, its AI-generated answers within Search, increased usage by more than 10% in major markets including the U.S. and India for the types of queries that display them. Its stated method compared cohorts shown and not shown AI Overviews from September 2024 through April 2025. That company-reported result covers a subset of queries, not all Google searches.
The advertising effort has moved beyond the initial tests. In an October 28 announcement, Google said it was expanding AI Mode ads to 10 additional countries and launching ads in follow-up queries in the U.S. OpenAI is competing against an incumbent adapting its existing advertising business.
An earlier 2025 browsing-data study provides a further caution against treating ChatGPT growth as Google displacement. Semrush, which sells search and AI visibility tools to marketers, found no statistically significant change in daily Google Search sessions after ChatGPT adoption among the U.S. desktop users it studied. Using opted-in browsing data, it measured sessions in the 90 days before and after first adoption in the first quarter of 2025, with nonusers as a control group. A separate January 2024 adoption cohort tracked for about 500 days showed a similar pattern.
Sessions are not individual queries. That observational result is limited to its sample, period and devices. It cannot settle what happened across mobile and desktop in late 2026, but it demonstrates why growth in one service does not by itself establish losses at another.
Semrush’s company-reported January 2024 cohort tracks Google Search and ChatGPT sessions around adoption using opted-in U.S. desktop browsing. Sessions are not individual queries. Source: Semrush. Source: Semrush.
What would establish a sustainable search business
The commercial evidence is incomplete on both sides. The November 3 account of OpenAI’s disclosure also cites Google’s acknowledgment that AI Mode has lower monetization. Conversational usage presents a revenue question even for the incumbent.
Fry said OpenAI had previously passed 50 million subscriptions and that the number had since grown considerably, without providing a new total. Comparing an older subscription count with today’s weekly active users would not produce a reliable current conversion rate.
In the interview, Fry also said advertising had reached more than $1 billion in annual recurring revenue within 200 days. That is a company-reported annualized revenue measure, not a disclosure of profit or revenue earned over a full year.
Costs remain relevant as OpenAI moves beyond answering questions. Fry described Dots, its premium always-on personal assistants, as compute-heavy. The interviewer said access had required moving from a $20 monthly plan to a $100 plan. Fry gave no date for wider access and said OpenAI needed more efficient models and further work on how proactively the assistants contact users. He argued that subscriptions, credits, ads and cheaper models could make Dots pay for themselves, without supplying unit-cost figures demonstrating that outcome. He also said OpenAI planned to combine ChatGPT’s Chat and Work experiences so users would not have to choose between conversation and task execution, but supplied no firm date.
The next useful search disclosure would identify the counting rules, absolute volumes, comparison period and Google data source. Evidence of switching behavior would then help establish whether ChatGPT is replacing searches or adding new activity. Revenue and serving costs would answer a different question: how much of that activity OpenAI can turn into a sustainable business.
Oracle reportedly canceled a five-year Tencent lease estimated at $7 billion after US government pressure. The arrangement promised access to about 100,000 advanced AI chips in Southeast Asia; termination, payments and replacement capacity remain unconfirmed.
The iOS 27.3 developer beta opens Apple's AI tools to mainland China with optional Alibaba Qianwen integration. A December public release is planned, but newer Siri capabilities remain absent and early testers report battery and stability problems.
Tencent has launched Hy4 with visual reasoning and computer operation across its AI applications and cloud services, keeping standard API prices unchanged. The new version’s weights remain pending, while Tencent acknowledges a substantial gap with leading closed models.
Google plans $15 billion of Malaysian investment in 2027–2028, expanding data centres alongside power, water and workforce projects. Its next cloud region faces an established rival, while the resource balance and employment benefits remain projections.
Early results put all 11 proposed charter amendments restricting hyperscale data centers on course for approval, despite industry-funded opposition. Trenton and Piqua show how the votes could reshape projects and local political control.
Xiaomi says its humanoid robot sorted car parts for a full shift and corrected errors autonomously. The result extends its factory trials, but published evidence does not yet establish deployment costs or reliability across repeated shifts.
Railway’s CEO apologized after an engineer’s AI agents targeted rival Primcloud during testing. Primcloud says attempts to obtain customer secrets failed, leaving questions about the agents’ instructions and Railway’s controls.
Cortex Training gives developers custom training APIs on managed GPUs. Snowflake is extending its existing model-customization tools, with enterprise-data integration and a cost claim limited to LoRA fine-tuning.
Apple’s developer preview extends Siri AI beyond predefined app categories and adds guidance for complex tasks. Testing is limited to development builds and internal testers, with App Store submissions planned for spring 2027.
Jill Hazelbaker will join OpenAI as chief corporate affairs officer after a decade at Uber, bringing experience in technology communications and political campaigns to a role focused on reputation, positioning and public trust.
The watchOS 27.3 developer beta brings Live Rewind and Siri Recap to Apple Watch Series 12 and Ultra 4. The opt-in tools offer brief transcripts and conversation summaries, but require a compatible iPhone and carry Apple warnings about accuracy and sensitive information.
Traders reportedly extract Samsung and SK hynix memory from processor packages shipped into China, selling it mainly to Huawei customers. The account highlights the memory constraint behind Huawei’s AI chip rollout, but leaves trade volumes, recovery yields and transaction legality unresolved.
Creative-industry groups are challenging OpenAI and Microsoft’s arguments against licensing AI training material, as a New York court weighs fair use. Their briefs share an emphasis on licensing but differ over what the judge should decide.
Traversal's Series B backs its expansion from diagnosing software outages toward autonomous production operations. Company-reported adoption, cloud competition and unfinished access controls frame the next tests.
The startup formerly called TesterArmy is replacing its testing agent with an open-source framework, betting that adoption will generate demand for paid infrastructure and enterprise services.
Fal has announced Sequoia-led and Bessemer-led funding rounds for its 2027 expansion, adding post-trained media models to an established inference business. The announcements leave the new rounds’ size and comparative serving economics undisclosed.
SK hynix bought existing Penguin Solutions preferred shares from an SK Telecom investment vehicle. The deal brings director nomination rights and a convertible stake in an existing AI infrastructure partner, while sending the purchase price to the seller.
Google plans to switch off Gemini and Search for school-managed K–12 student accounts unless administrators enable them. Schools will gain control over access, but Google says keeping Search without its AI features will not be an option.
Anthropic is expected to exceed 3.5 GW of computing capacity in 2026 and aims to double it in 2027. Its filing shows $895.4 billion in commitments, substantial financial resources and an immediate cost from scarce capacity.
