OpenAI has started testing a new format of sponsored advertisements that include images inside ChatGPT, with initial trials launched in the United States, according to The Verge. The company says these image-based ads will first appear only during image-generation interactions.
Background
A few months ago OpenAI began testing advertisements in ChatGPT; in Hungary, ads have been appearing in conversations with the chatbot since August. The previous ad format was restrained: typically only the sponsor’s name, logo and a link were displayed under a clearly marked “sponsored” section, separated from the chatbot’s normal responses.
What’s changing
The new tests introduce sponsored content that can include photos, but OpenAI asserts that the ad images will not influence the images or other outputs that ChatGPT generates. According to the company, the ads will be visually distinct from the user-requested AI images, as noted by The Verge and illustrated in related coverage.
Who is exempt
Users with ChatGPT Plus, Pro or enterprise subscriptions will not see the new types of ads. OpenAI also states that the image-based sponsored content will not appear in conversations about sensitive topics.
Known and unknowns
The trial is currently limited to the United States, and OpenAI emphasizes that the new ad format will not alter generated content. The company has not provided a timeline for when these image-based sponsored ads might be rolled out to all users.
Why this matters
The move underscores OpenAI’s continued experimentation with monetization within ChatGPT while attempting to preserve user experience: by exempting paying subscribers and avoiding sensitive topics, the company aims to limit disruption. Given the lack of a broader rollout schedule, users and industry observers should expect further updates as testing progresses.
