Advertising Is Coming to AI Chatbots and Could Shape Their Answers

Summary

As people increasingly use AI conversations to search for information and make decisions, the article examines how advertising could affect the answers they receive. Unlike banner ads or sponsored links, advertising in chatbots may be integrated into the conversation itself. OpenAI has introduced ads designed around ChatGPT interactions, while Google is testing sponsored answers in Gemini; both approaches appear connected to real-time bidding and behavioral micro-targeting. AI companies say advertising will not change the substance of answers, but the article argues that advertising technology is difficult for outsiders to audit and that safeguards currently rely largely on providers honoring voluntary promises. Perplexity tested sponsored conversations and later discontinued the program, while clear labels may still not help users distinguish paid material from organically generated information inside a dialogue. The authors describe a possible “zero-click internet” in which purchases happen inside corporate-controlled AI environments, leaving providers in control of user attention and commercial opportunities. They warn that generated answers can compress competing perspectives, reduce the visibility of minority or dissenting views, and create pay-to-play advantages for large companies. The article also links targeted AI advertising to misinformation and dark-money election influence, noting that personalized and ephemeral falsehoods may escape public scrutiny and fact-checking. Existing defenses, including the European Union’s Digital Services Act, were designed mainly for social media and search engines and do not guarantee the accuracy of each chatbot response. The authors call for policymakers to prevent sponsors from shaping the evidence and substance of AI conversations, arguing that labels alone are unlikely to be sufficient.