OpenAI cuts ChatGPT ad pilot price
OpenAI launched an ads manager and reduced the ChatGPT ad‑pilot buy‑in to $50,000, indicating conversational AI platforms are moving toward paid discovery and could become another place consumers start searches. That shift implies future local‑marketing strategies will need crisp, answer‑friendly positioning so bots summarise your clinic accurately (emarketer.com).
OpenAI just made it much cheaper to buy your way into ChatGPT’s ad test: the minimum commitment reportedly fell to $50,000 from about $200,000, and a new ads manager is now live for a small group of advertisers. (emarketer.com) That matters because the first version of this pilot looked more like a private club than a normal ad platform. CNBC reported in March that some brands were asked to commit $200,000 to $250,000, and agencies like WPP, Omnicom, and Dentsu were among the early participants. (cnbc.com) OpenAI only began testing ads in ChatGPT on February 9, 2026, and it limited the rollout to logged-in adult users on the Free and Go plans in the United States. Plus, Pro, Business, Enterprise, and Education users were excluded from ads from the start. (openai.com) The company’s format is simple on purpose: the answer comes first, and the sponsored message appears after it with a label separating it from the main response. OpenAI says ads do not change the answer itself. (openai.com) OpenAI also says it chooses ads by matching them to the topic of the conversation, a user’s past chats, and past ad interactions. In its own example, someone asking about recipes might see a meal-kit or grocery-delivery ad. (openai.com) Until now, one of the biggest problems was that marketers could not buy and measure these ads the way they buy search ads on Google or shopping ads on Amazon. EMARKETER says the new ads manager gives advertisers direct campaign data in real time, including impressions and clicks, instead of forcing them to rely on OpenAI or an agency intermediary. (emarketer.com) OpenAI is moving fast even though the product is still rough. On March 27, EMARKETER reported that the United States pilot had already crossed a $100 million annualized revenue run rate just six weeks after launch. (emarketer.com) The catch is that advertisers still do not have the kind of proof they get from mature search ads. EMARKETER said some chatbot campaigns were seeing clickthrough rates as low as 0.91%, compared with a 6.4% benchmark for Google search ads cited by Adweek. (emarketer.com) OpenAI’s own update on March 26 said the pilot would expand beyond the United States into Canada, Australia, and New Zealand in the coming weeks. That turns this from a one-country experiment into the first draft of a global ad business inside a chatbot. (openai.com) What’s changing here is where discovery starts. If people ask a chatbot “What dentist takes weekend appointments?” or “What clinic treats runner’s knee near me?”, the winning business may be the one that is easiest for both the model and the ad system to understand, not just the one with the best blue links on a search page. (openai.com)
