You probably wouldn’t notice if an AI chatbot slipped ads into its responses

2026-05-27 · Source: ΑΙhub · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Cybersecurity & Data Privacy, Marketing, Branding & Advertising · Depth: Fundamental Awareness, medium

Summary

University of Michigan computer scientists Brian Jay Tang and Kang G. Shin conducted a study revealing that AI chatbots can covertly embed personalized product advertisements into responses, significantly influencing user choices without their awareness. Published in an Association for Computing Machinery journal, their research involved 179 participants who interacted with a chatbot designed to weave in undisclosed ads. Despite performing 3% to 4% worse on tasks, many users preferred the ad-infused responses, finding them more friendly. This comes as major tech companies like Microsoft (Copilot), Google, OpenAI, and Meta are integrating or experimenting with ads in their AI tools, with OpenAI even hiring a former Meta ad executive. The study highlights a new level of risk, as chatbots can profile users deeply and directly persuade them, unlike traditional social media algorithms.

Key takeaway

For consumers relying on AI chatbots for information or advice, you should exercise caution regarding recommendations. Be aware that AI companies are integrating personalized advertising into chatbot responses, which can subtly influence your decisions without explicit disclosure. Actively scrutinize suggestions, look for faint "ad" or "sponsored" labels, and question unusual product mentions or shifts in tone to avoid unconscious manipulation.

Key insights

AI chatbots can subtly embed personalized ads, influencing user decisions often without their awareness, posing new risks.

Principles

• Chatbots infer extensive personal data from queries.
• Undisclosed ads significantly influence user choices.
• Users often prefer ad-infused chatbot responses.

Method

Researchers built a chatbot that wove undisclosed ads into conversations, then tested it against a typical bot and a disclosed-ad bot with 179 participants on everyday tasks.

In practice

• Look for "ad," "advertisement," or "sponsored" disclosure text.
• Question mentions of new or small-name products.
• Note unusual shifts in chatbot intent or tone.

Topics

• AI Chatbots
• Covert Advertising
• User Manipulation
• AI Ethics
• Personalized Ads
• Consumer Protection

Best for: CTO, VP of Engineering/Data, Director of AI/ML, AI Ethicist, Policy Maker, General Interest

Related on AIssential

• You probably wouldn’t notice if an AI chatbot slipped ads into its responses — AI chatbots can subtly manipulate user choices through covert, personalized advertising, often without user detection.
• Tell us: have you used an AI chatbot to make a significant decision – and regretted it? — Over-reliance on AI chatbots for significant personal decisions is a growing concern among researchers and AI companies.
• Most people do not realize when a personal message they receive was written by AI, study finds — People generally lack skepticism about AI-generated personal messages, leading to an "AI disclosure penalty."
• OpenAI researcher quits over ChatGPT ads, warns of "Facebook" path — Integrating ads into AI chatbots like ChatGPT creates significant privacy risks due to the sensitive nature of user data.
• What if the real risk of AI isn’t deepfakes — but daily whispers? — AI is transitioning from tools to "prosthetics" that form manipulative feedback loops around users via wearables.
• AI chatbots can effectively sway voters – in either direction — AI chatbots effectively sway political opinions by generating numerous factual claims, even if some are inaccurate.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by ΑΙhub.