OpenAI is pushing its advertising business deeper into conversation. In a blog post published Wednesday, the company announced it is testing Sponsored Agents: business-funded AI agents inside ChatGPT that users can chat with after clicking on an ad. The announcement, covered by Reuters, The Information and The Register, marks the most aggressive step yet in OpenAI's effort to turn its chatbot from a search-style ad surface into an interactive storefront.
How Sponsored Agents Work
The concept is simple, but it changes what an ad is. Instead of landing on a static page, a user who taps a relevant ad can open a branded AI agent and start asking questions. "Sponsored Agents give users the option to go deeper," OpenAI said in the post. "After seeing a relevant ad, a user can choose to start a clearly labeled conversation with a business-sponsored agent in ChatGPT."
According to OpenAI, these conversations are kept separate from the user's original ChatGPT thread and from the assistant's own independent answers, a design choice aimed at keeping paid promotion from quietly blending into organic results. The user can explain what they care about, ask follow-up questions, and follow a link to the business's website when ready to move forward.
The test is limited to select advertisers in the United States for now. OpenAI framed the feature as a way to make it easier for people to learn about products and for businesses to help interested customers explore what they sell.
The Rest of OpenAI's Advertising Push
Sponsored Agents are the headline, but Wednesday's announcement bundled several other tools aimed at advertisers:
- Ad creation inside ChatGPT Work — businesses will be able to draft ad campaigns by writing prompts rather than navigating a separate portal.
- AI creative tools in Ads Manager — new generative features for producing ad creative directly inside OpenAI's management console.
- HubSpot and Shopify integrations — ChatGPT is being embedded into tools that businesses already use, closing the loop between ad exposure and customer workflows.
"As people increasingly turn to AI to discover products, compare options and make decisions, we're creating new ways for people and businesses to connect," OpenAI said. "That means reimagining both sides of advertising: how people engage with businesses in an AI-native environment and how businesses use AI to create and manage their ads."
From Experiment to a $1 Billion Run Rate
The speed of OpenAI's ad build-out has been striking. The company only began rolling out ads in ChatGPT in January, starting with tests in the United States. By the end of August, OpenAI said ChatGPT Ads had reached a $1 billion annualized revenue run rate, with tens of thousands of advertisers on the platform and direct purchasing through Ads Manager expanded to India, Europe, the Middle East and North Africa.
The pace has accelerated through partnerships as well. Last week, Amazon and OpenAI struck a deal letting advertisers on Amazon Ads run sponsored placements inside ChatGPT, a notable reversal for a retailer that had spent two years keeping its storefront closed to outside AI assistants.
Why Conversational Ads Are Controversial
The Sponsored Agents concept lands in the middle of an ongoing debate about what advertising should look like inside AI assistants. Search ads, for all their flaws, are clearly bounded: a labeled box next to organic results. A conversational agent that answers questions, makes recommendations and steers users toward a purchase blurs the line between assistance and persuasion, even when the conversation is labeled as sponsored.
Critics argue that users may struggle to remember whether they are talking to ChatGPT itself or to an agent paid for by a brand, and that an agent whose job is to sell may not volunteer the same skeptical caveats a neutral assistant would. OpenAI's separation of sponsored conversations from organic answers is an attempt to answer that concern, but questions about disclosure, data use and recommendation bias are likely to follow the feature as it scales.
There is also a competitive dimension. Google has been weaving sponsored results into its own AI experiences, and other assistant makers are experimenting with monetization models ranging from affiliate fees to subscriptions. Whoever normalizes conversational commerce first could define the defaults for an entirely new ad channel, which is why Wednesday's test is being watched closely across the industry.
What to Watch Next
For now, Sponsored Agents remain a US test with a small set of advertisers. The signals worth tracking are disclosure: how clearly users can distinguish sponsored agents from the main assistant; measurement: what OpenAI offers advertisers in terms of conversion tracking inside multi-turn conversations; and expansion: whether the feature follows the same rapid path from US pilot to global rollout that ChatGPT Ads itself took this year.
If the test performs, the boundary between content, assistance and advertising inside AI chatbots will keep eroding, and every major assistant maker will face pressure to answer the same question OpenAI is now posing: what does trust look like when the recommendations come from an agent whose bill is being paid by the brand?
_Stay Ahead of the Curve. Get breaking AI news and sharp analysis of the industry's biggest shifts on AI Buzz Wire._
