ChatGPT Ads Are Coming! Are We Ready?

And it happened! OpenAI officially announced it has started testing ads for its AI chatbot, ChatGPT. This development could be a turning point for the digital marketing and advertising ecosystem. So, what does this new ad model promise, and what opportunities and threats does it present? Let’s dig in.

According to OpenAI’s statements, ads will initially be shown only to users in the free and low-cost “Go” ($8/month) tiers in the U.S. Those with paid subscriptions like Plus, Pro, and Enterprise will not see these ads. This strategy shows OpenAI’s effort to cover its massive operational costs on one hand, and its desire to protect the paid user experience on the other. The company emphasizes that they prioritize user trust above all, that ads will not influence the model’s answers, and that the privacy of conversations will not be shared with advertisers.

The leaked control panel images and official statements about how the ads will work signal a well-thought-out system. Ads will appear not within the chat interface, but below it, in a separate section clearly labeled as “sponsored.” This is an approach reminiscent of Google’s ad placement in search results, aiming not to mislead the user. Most importantly, users will have full control over ad personalization. They can view their ad history, delete interest inferences, and, if they wish, turn off personalization completely. When personalization is off, ads will be shown based only on the current conversation’s context. This is a significant assurance for privacy-conscious users.

Ad targeting will be based on conversation topics, past chats, and previous ad interactions. For example, a user chatting about recipes might see an ad for a grocery delivery service. OpenAI states that this model promises a future where small and medium-sized businesses can run campaigns by describing their goals in natural language, rather than dealing with complex ad panels. This could be an exciting step towards the democratization of advertising.

However, there’s another side to the coin. The Super Bowl ads by rival AI company Anthropic, which mocked ChatGPT’s ad model, also summarize the industry’s skepticism. Ads, no matter how well-intentioned, always have the potential to negatively affect the AI’s impartiality and the user experience. As technical SEO consultant Jono Alderson pointed out, when you create a second, machine-readable version of a page (like markdown), you are essentially creating a second candidate for reality. How will the system know which version is the truth? This also brings up “cloaking” concerns. SEO consultant David McSweeney warns that Cloudflare’s new Markdown for Agents feature could create a “shadow web” by making it easier for sites to show different content to bots than to humans.

So, what should we do? This development is one of the clearest signs of how search and discovery behaviors are evolving. We now need to be present not just on Google, but also in AI chat interfaces, positioning our brand and products in the right context. This new platform has the potential to understand user intent much more deeply and to appear at the exact moment of need. At Stradiji, we are closely following these developments and already modeling new opportunities and strategies for our clients. Remember, change is inevitable. What matters is how quickly and intelligently we adapt to it.