Summary: ChatGPT began serving ads in the United States on February 9, 2026 to logged-in adults on the Free and Go tiers, while Plus, Pro, Business, Enterprise, and Education plans remained ad-free. By August 11, the test had reached Canada, Australia, New Zealand, the United Kingdom, Mexico, Brazil, Japan, and South Korea. OpenAI reported no impact on trust metrics, but free users who turn off personalized ads receive a smaller daily message allowance.
ChatGPT’s ad test went live in the United States on February 9, 2026. OpenAI drew the boundaries narrowly: only logged-in adults on the Free and Go tiers were included, while Plus, Pro, Business, Enterprise, and Education plans are kept free of ads. By August 11 the test had reached the UK, Mexico, Brazil, Japan, and South Korea, following Canada, Australia, and New Zealand, which came aboard back in March. One country to eight markets inside six months — that is a product line, not an experiment.
How the Ads Look and How They’re Picked
Every ad carries a “sponsored” label and sits visually apart from the conversational answer, and OpenAI says outright that ads “do not influence” responses. Picking one draws on three signals at once: the topic of the conversation at hand, the content of earlier chats, and how the user has interacted with ads before.
The privacy boundaries are drawn tighter than on most platforms. Advertisers receive aggregate-level reporting only, never individual identities. Minors — self-reported or predicted alike — are excluded outright. And sensitive subjects such as health, mental health, and politics get no ads anywhere near them.
What Opting Out Costs You
Users on the free tier may switch personalized ads off, but OpenAI lays out the trade: opting out shrinks the daily free message allowance. Watching ads is, quite directly, how the free tier gets paid for. The remaining controls are thorough — report an ad, send feedback, check “why am I seeing this ad,” wipe ad interaction data with one tap, and browse ads history.
Early Results and the Pace of Expansion
Early figures arrived with the March 26 update: ads produced “no impact” on consumer trust metrics, and dismissal rates held low — the evidence OpenAI pointed to when it launched international expansion. In May came the announced plan covering the UK and four more markets; on August 11 all five were confirmed live, with further markets promised before the year is out. On the business side, openai.com/advertisers opened for business. The long-term direction is on record as well: ad formats, objectives, and buying models will keep evolving.
Attention Economics Inside a Chat Interface
Three lessons from these six months stand out for product people. First, the keyword matching of search advertising becomes “conversational intent matching” once inside a chat interface — intent signals outpower keywords, which is at once the genuine selling point of AI-era ad products and the heart of the privacy argument. Second, “pay to remove ads” tiering has settled into shape: free and low-cost tiers carry ads while premium tiers stay clean — a structure plenty of products will copy. Third, trust metrics set the ceiling on any ads product; that OpenAI is willing to cite “no impact on trust” publicly shows it understands users will defect the instant ads degrade answer quality. Turning the free tier ad-supported is at bottom a trade of user trust for scale — a balance every team that monetizes through advertising should be watching.