Business Insider’s report that ChatGPT’s advertiser count rose from roughly 300 in April to more than 820 in July marks a sharper change for users than the raw number suggests: OpenAI is now scaling ads inside both the free product and the $8-per-month ChatGPT Go tier, not treating advertising as a small free-tier experiment.
Sensor Tower, the mobile-intelligence firm cited by Business Insider, estimated that hourly ad impressions in the ChatGPT app roughly doubled from April to July and that at least 160 advertisers arrived during July alone. OpenAI has not published an advertiser roster, impression volume, or revenue breakdown that independently confirms those figures. But the company’s own product updates show that the machinery for a broader ad business is in place: a self-service Ads Manager, cost-per-click buying, conversion reporting, and a relevance-based auction.
For Windows users who use ChatGPT as a research assistant, a troubleshooting tool, or an alternative to conventional web search, the practical change is straightforward. The sponsored result may sit outside the answer, but OpenAI is matching it against the subject and apparent intent of the conversation. The company says advertisers cannot alter the response itself; it does not say conversations are irrelevant to advertising. They are central to deciding which ad is displayed.
The most eye-catching claim in the report is the move past 820 advertisers. It should be read as Sensor Tower’s observed or modeled estimate, rather than as an OpenAI disclosure. Neither the original reporting nor OpenAI has specified whether this is a count of unique legal advertisers, brand names, active campaigns, or buyers appearing within a defined geography and time window.
That distinction is important because an emerging ad market can look substantially larger when one agency runs campaigns for many brands, when a retailer buys several campaigns, or when brands appear briefly during a test. OpenAI’s public advertiser materials describe a beta Ads Manager and agency-channel buying, but do not publish active advertiser totals.
Still, the named brands change the character of the market. Early ChatGPT advertising was visibly populated by software, productivity and lead-generation companies such as Speechify, Jotform and Natural Intelligence. The latest reported mix includes Booking Holdings, Intuit, Home Depot and L.L.Bean — businesses that sell travel, tax and finance services, home-improvement goods, and consumer products.
That is evidence of a move from a technology-sector trial toward mainstream performance advertising: ads placed where a user may be comparing products, planning a trip, looking for a service, or trying to complete a purchase. It also explains why shopping remains the largest category in Sensor Tower’s analysis. ChatGPT conversations tend to contain more explicit intent than social-media scrolling, but less neatly packaged intent than a conventional search query.
OpenAI’s May announcement removed a major barrier to expanding the buyer pool. During the initial pilot, advertisers reportedly needed substantial commitments, while the new self-service system enables companies to register, supply creative, choose bids and budgets, and monitor campaigns. OpenAI added cost-per-click buying alongside impression-based buying, with advertiser reporting for impressions, clicks, click-through rate, spending, effective rates and conversions.
The rapid increase in advertiser numbers therefore has a mundane, credible explanation: OpenAI opened the door to smaller buyers and supplied the basic tools they expect. It does not, by itself, prove that ChatGPT ads produce repeatable sales.
OpenAI introduced Go in the United States at $8 per month in January, positioning it between the free version and the higher-priced Plus and Pro subscriptions. Go adds more access to popular ChatGPT features, including messaging, image creation, file uploads and memory, while retaining advertising. The free tier also carries ads. Plus, Pro, Business, Enterprise and Edu accounts do not.
In other words, paying for ChatGPT no longer automatically means paying to remove commercial content. For individuals and small teams trying to control software spending, Go is effectively an ad-supported subscription — a model familiar from streaming, but more sensitive in an assistant that users consult about work, money, health, education and personal decisions.
OpenAI says ads do not appear for users who state that they are under 18 or whom the company predicts are under 18. Its help documentation also says logged-out users can see age-appropriate ads, because OpenAI has less age information for those sessions. That creates a wider potential inventory pool than a logged-in-adults-only approach, even as the company says it is taking a phased rollout.
The geographic scope is also more limited than a headline about a billion monthly users might imply. OpenAI says ads are currently rolled out in the United States, United Kingdom, Australia, New Zealand and Canada, with availability still subject to change as testing expands. Sensor Tower’s one-billion figure refers to monthly active users of the ChatGPT mobile app globally in May, not necessarily to one billion people in advertising-eligible markets.
That gap is the first material constraint on the revenue story. OpenAI has global reach, but its ad inventory, policies and buying tools are rolling out market by market.
That separation is meaningful. An ad card beneath an answer is easier for a user to identify and ignore than a recommendation woven invisibly into ChatGPT’s prose. OpenAI also says an advertisement is not an endorsement of the advertiser or its product.
But answer independence should not be confused with context independence. OpenAI’s advertiser documentation says ad delivery uses the current conversation’s context and intent, the advertiser’s landing page and creative, advertiser-provided context hints, and — where personalized ads are enabled — selected signals from a user’s broader ChatGPT experience. Advertisers do not receive chat histories, memories, or personal details; OpenAI says they receive only aggregate performance data. The matching happens inside OpenAI’s system.
For a user, that means a query such as “compare Windows backup services,” “best software for filing estimated taxes,” or “cheap flights to Chicago” can establish the context for an ad underneath the answer. It is a more intimate signal than a keyword typed into a search engine because it may include the constraints and reasons behind the request.
OpenAI offers controls, but they need to be understood precisely. Turning off personalized ads does not remove ads. It makes them less tailored by preventing broader ChatGPT activity from informing ad relevance. Disabling use of past chats and memory means ad selection relies more heavily on the active conversation. Users can also hide or report a specific ad, and OpenAI says there will always be an ad-free paid option.
For organizations, the cleanest boundary remains a Business, Enterprise or Edu account. Those plans are ad-free under OpenAI’s published policy, though IT administrators should still distinguish that commercial-policy promise from their own data-governance, retention and approved-AI-use requirements.
OpenAI’s current ad policies place financial services, healthcare and legal services in a restricted bucket. They may be approved on a gradual, case-by-case basis; other categories are initially focused on consumer areas such as household goods, local services, travel, digital products and education. Ads are barred around sensitive contexts, including personal and mental health discussions, political content, vulnerable interactions, harmful activity, self-harm, regulated goods and privacy-related conversations.
Those safeguards address where an ad can appear. They do not settle the harder user-experience issue: whether an ad below a conversational answer feels like a recommendation when the assistant has just helped frame the question. OpenAI’s labeling and response separation are necessary protections, but their effectiveness will depend on the prominence of disclosures and the consistency of enforcement as inventory grows.
The other unresolved issue is measurement. OpenAI’s Ads Manager reports standard platform metrics and conversion data, but Digiday reported in June that OpenAI had not yet implemented third-party verification to confirm viewability, human reach and brand-safe delivery. Dave Dugan, OpenAI’s head of global ad solutions, told the publication that outside measurement was a natural next step, without naming a partner or timetable.
That is a business problem as much as an advertiser problem. The move from 300 to more than 820 advertisers can be driven by novelty, inexpensive test budgets and the availability of self-service buying. Sustained spending will depend on whether OpenAI can prove that a click below a generated answer produces outcomes advertisers trust.
ChatGPT’s billion-mobile-user milestone gives OpenAI the audience to build a serious advertising business. The immediate consequence for users is more concrete: Free and Go conversations are now becoming a commercial surface, and the settings that limit personalization do not remove the ads themselves.
For Windows users who use ChatGPT as a research assistant, a troubleshooting tool, or an alternative to conventional web search, the practical change is straightforward. The sponsored result may sit outside the answer, but OpenAI is matching it against the subject and apparent intent of the conversation. The company says advertisers cannot alter the response itself; it does not say conversations are irrelevant to advertising. They are central to deciding which ad is displayed.
The 820 figure is an estimate, not OpenAI’s official count
The most eye-catching claim in the report is the move past 820 advertisers. It should be read as Sensor Tower’s observed or modeled estimate, rather than as an OpenAI disclosure. Neither the original reporting nor OpenAI has specified whether this is a count of unique legal advertisers, brand names, active campaigns, or buyers appearing within a defined geography and time window.That distinction is important because an emerging ad market can look substantially larger when one agency runs campaigns for many brands, when a retailer buys several campaigns, or when brands appear briefly during a test. OpenAI’s public advertiser materials describe a beta Ads Manager and agency-channel buying, but do not publish active advertiser totals.
Still, the named brands change the character of the market. Early ChatGPT advertising was visibly populated by software, productivity and lead-generation companies such as Speechify, Jotform and Natural Intelligence. The latest reported mix includes Booking Holdings, Intuit, Home Depot and L.L.Bean — businesses that sell travel, tax and finance services, home-improvement goods, and consumer products.
That is evidence of a move from a technology-sector trial toward mainstream performance advertising: ads placed where a user may be comparing products, planning a trip, looking for a service, or trying to complete a purchase. It also explains why shopping remains the largest category in Sensor Tower’s analysis. ChatGPT conversations tend to contain more explicit intent than social-media scrolling, but less neatly packaged intent than a conventional search query.
OpenAI’s May announcement removed a major barrier to expanding the buyer pool. During the initial pilot, advertisers reportedly needed substantial commitments, while the new self-service system enables companies to register, supply creative, choose bids and budgets, and monitor campaigns. OpenAI added cost-per-click buying alongside impression-based buying, with advertiser reporting for impressions, clicks, click-through rate, spending, effective rates and conversions.
The rapid increase in advertiser numbers therefore has a mundane, credible explanation: OpenAI opened the door to smaller buyers and supplied the basic tools they expect. It does not, by itself, prove that ChatGPT ads produce repeatable sales.
Ads are reaching paid users who chose the lower-cost plan
The submitted report accurately identifies the $8 plan, but it leaves out a consequential detail: this is ChatGPT Go, OpenAI’s lower-cost subscription tier, and ads are an explicit part of its trade-off.OpenAI introduced Go in the United States at $8 per month in January, positioning it between the free version and the higher-priced Plus and Pro subscriptions. Go adds more access to popular ChatGPT features, including messaging, image creation, file uploads and memory, while retaining advertising. The free tier also carries ads. Plus, Pro, Business, Enterprise and Edu accounts do not.
In other words, paying for ChatGPT no longer automatically means paying to remove commercial content. For individuals and small teams trying to control software spending, Go is effectively an ad-supported subscription — a model familiar from streaming, but more sensitive in an assistant that users consult about work, money, health, education and personal decisions.
OpenAI says ads do not appear for users who state that they are under 18 or whom the company predicts are under 18. Its help documentation also says logged-out users can see age-appropriate ads, because OpenAI has less age information for those sessions. That creates a wider potential inventory pool than a logged-in-adults-only approach, even as the company says it is taking a phased rollout.
The geographic scope is also more limited than a headline about a billion monthly users might imply. OpenAI says ads are currently rolled out in the United States, United Kingdom, Australia, New Zealand and Canada, with availability still subject to change as testing expands. Sensor Tower’s one-billion figure refers to monthly active users of the ChatGPT mobile app globally in May, not necessarily to one billion people in advertising-eligible markets.
That gap is the first material constraint on the revenue story. OpenAI has global reach, but its ad inventory, policies and buying tools are rolling out market by market.
The answer may stay independent, but the conversation drives the ad match
OpenAI’s core promise is that paid placements will not influence ChatGPT’s answers. Its help center says the ad system is separate from the chat model and that advertisers cannot shape, rank or alter a response. Ads appear below the end of a response, are labeled as sponsored, and include an advertiser name, logo, headline, description, image, and destination page.That separation is meaningful. An ad card beneath an answer is easier for a user to identify and ignore than a recommendation woven invisibly into ChatGPT’s prose. OpenAI also says an advertisement is not an endorsement of the advertiser or its product.
But answer independence should not be confused with context independence. OpenAI’s advertiser documentation says ad delivery uses the current conversation’s context and intent, the advertiser’s landing page and creative, advertiser-provided context hints, and — where personalized ads are enabled — selected signals from a user’s broader ChatGPT experience. Advertisers do not receive chat histories, memories, or personal details; OpenAI says they receive only aggregate performance data. The matching happens inside OpenAI’s system.
For a user, that means a query such as “compare Windows backup services,” “best software for filing estimated taxes,” or “cheap flights to Chicago” can establish the context for an ad underneath the answer. It is a more intimate signal than a keyword typed into a search engine because it may include the constraints and reasons behind the request.
OpenAI offers controls, but they need to be understood precisely. Turning off personalized ads does not remove ads. It makes them less tailored by preventing broader ChatGPT activity from informing ad relevance. Disabling use of past chats and memory means ad selection relies more heavily on the active conversation. Users can also hide or report a specific ad, and OpenAI says there will always be an ad-free paid option.
For organizations, the cleanest boundary remains a Business, Enterprise or Edu account. Those plans are ad-free under OpenAI’s published policy, though IT administrators should still distinguish that commercial-policy promise from their own data-governance, retention and approved-AI-use requirements.
Financial-services growth tests OpenAI’s safety rules
Sensor Tower’s reporting says financial-services ads increased from 2% of ChatGPT ad spending in April to 12% in July, the fastest growth among categories. That shift is notable because financial products are precisely where conversational context can create a perception of advice, endorsement or individualized suitability even when the ad is visually separate from the answer.OpenAI’s current ad policies place financial services, healthcare and legal services in a restricted bucket. They may be approved on a gradual, case-by-case basis; other categories are initially focused on consumer areas such as household goods, local services, travel, digital products and education. Ads are barred around sensitive contexts, including personal and mental health discussions, political content, vulnerable interactions, harmful activity, self-harm, regulated goods and privacy-related conversations.
Those safeguards address where an ad can appear. They do not settle the harder user-experience issue: whether an ad below a conversational answer feels like a recommendation when the assistant has just helped frame the question. OpenAI’s labeling and response separation are necessary protections, but their effectiveness will depend on the prominence of disclosures and the consistency of enforcement as inventory grows.
The other unresolved issue is measurement. OpenAI’s Ads Manager reports standard platform metrics and conversion data, but Digiday reported in June that OpenAI had not yet implemented third-party verification to confirm viewability, human reach and brand-safe delivery. Dave Dugan, OpenAI’s head of global ad solutions, told the publication that outside measurement was a natural next step, without naming a partner or timetable.
That is a business problem as much as an advertiser problem. The move from 300 to more than 820 advertisers can be driven by novelty, inexpensive test budgets and the availability of self-service buying. Sustained spending will depend on whether OpenAI can prove that a click below a generated answer produces outcomes advertisers trust.
ChatGPT’s billion-mobile-user milestone gives OpenAI the audience to build a serious advertising business. The immediate consequence for users is more concrete: Free and Go conversations are now becoming a commercial surface, and the settings that limit personalization do not remove the ads themselves.
References
- Primary source: 디지털투데이
Published: 2026-08-03T05:13:55+00:00
Loading…
www.digitaltoday.co.kr - Related coverage: openai.com
New ways to buy ChatGPT ads | OpenAI
OpenAI expands ChatGPT ads with a beta self-serve Ads Manager, CPC bidding, and enhanced measurement tools—built to protect privacy and keep conversations separate from ads.openai.com - Related coverage: help.openai.com
Loading…
help.openai.com - Related coverage: digiday.com
Loading…
digiday.com - Related coverage: exame.com
Loading…
exame.com - Related coverage: adexchanger.com
Loading…
www.adexchanger.com - Related coverage: skift.com
Loading…
skift.com - Related coverage: digiday.com
Loading…
digiday.com - Related coverage: greghal.no
Loading…
greghal.no - Related coverage: help.openai.com
Loading…
help.openai.com - Related coverage: theinformation.com
Loading…
www.theinformation.com - Related coverage: windowscentral.com
Loading…
www.windowscentral.com - Related coverage: openai.com
Loading…
openai.com - Related coverage: seerinteractive.com
Loading…
www.seerinteractive.com - Related coverage: adthena.com
Loading…
www.adthena.com - Related coverage: chatgptadlibrary.com
Loading…
www.chatgptadlibrary.com - Related coverage: cloro.dev
Loading…
cloro.dev - Related coverage: cloro.dev
Loading…
cloro.dev - Related coverage: intuitionlabs.ai
- Related coverage: adthena.com
Loading…
www.adthena.com - Related coverage: help-lb.openai.com
Loading…
help-lb.openai.com - Related coverage: cdn.openai.com
Loading…
cdn.openai.com - Related coverage: axios.com
OpenAI launches self-serve ad platform
The new platform makes it easier for a broader set of advertisers to buy ads on ChatGPT.www.axios.com
- Related coverage: techradar.com
'We are growing revenue four times faster than the companies who defined the Internet and mobile eras': OpenAI says it's making $2 billion a month — mostly from enterprise users | TechRadar
OpenAI says it is now making $2 billion in monthly revenuewww.techradar.com