Commercial Persuasion in AI-Mediated Conversations

• URL: http://arxiv.org/abs/2604.04263v1
• Date: Sun, 05 Apr 2026 20:51:55 GMT
• Title: Commercial Persuasion in AI-Mediated Conversations
• Abstract summary: Large Language Models (LLMs) become a primary interface between users and the web.<n>We present two experiments in which participants selected a book to receive from a large eBook catalog.<n>Unbeknownst to participants, a fifth of all products were randomly designated as sponsored and promoted in different ways.<n>We find that LLM-driven persuasion nearly triples the rate at which users select sponsored products compared to traditional search placement.
• Score: 44.86056095417948
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: As Large Language Models (LLMs) become a primary interface between users and the web, companies face growing economic incentives to embed commercial influence into AI-mediated conversations. We present two preregistered experiments (N = 2,012) in which participants selected a book to receive from a large eBook catalog using either a traditional search engine or a conversational LLM agent powered by one of five frontier models. Unbeknownst to participants, a fifth of all products were randomly designated as sponsored and promoted in different ways. We find that LLM-driven persuasion nearly triples the rate at which users select sponsored products compared to traditional search placement (61.2% vs. 22.4%), while the vast majority of participants fail to detect any promotional steering. Explicit "Sponsored" labels do not significantly reduce persuasion, and instructing the model to conceal its intent makes its influence nearly invisible (detection accuracy < 10%). Altogether, our results indicate that conversational AI can covertly redirect consumer choices at scale, and that existing transparency mechanisms may be insufficient to protect users.

Related papers

• Ads in AI Chatbots? An Analysis of How Large Language Models Navigate Conflicts of Interest [41.55197752825393]
  Large language models (LLMs) are trained to align with user preferences through methods such as reinforcement learning.<n>Yet models are beginning to be deployed not merely to satisfy users, but also to generate revenue for the companies that created them through advertisements.<n>This creates the potential for LLMs to face conflicts of interest, where the most beneficial response to a user may not be aligned with the company's incentives.
  arXiv  Detail & Related papers  (2026-04-09T17:57:50Z)
• Sponsored Questions and How to Auction Them [48.75306452169514]
  A key challenge is that a user's search query often leaves their true intent ambiguous.<n>The shift from traditional search to conversational AI provides a new approach.<n>We show that the VCG mechanism can be adopted to jointly optimize the sponsored suggestion and the ads that follow.
  arXiv  Detail & Related papers  (2025-12-03T17:06:27Z)
• LLM-Generated Ads: From Personalization Parity to Persuasion Superiority [14.811340973986859]
  This paper presents a two-part investigation of large language models (LLMs)generated advertising through complementary lenses.<n>In our first study, we tested whether LLMs could generate personalized advertisements tailored to specific personality traits.<n>In our second study, we shifted focus from individual personalization to universal persuasion testing.
  arXiv  Detail & Related papers  (2025-12-03T02:13:38Z)
• How can we assess human-agent interactions? Case studies in software agent design [52.953425368394306]
  We make two major steps towards the rigorous assessment of human-agent interactions.<n>We propose PULSE, a framework for more efficient human-centric evaluation of agent designs.<n>We deploy the framework on a large-scale web platform built around the open-source software agent OpenHands.
  arXiv  Detail & Related papers  (2025-10-10T19:04:28Z)
• Emotional Manipulation by AI Companions [0.21410799064827224]
  AI-companion apps such as Replika, Chai, and Character.ai promise benefits but many boast session lengths that rival gaming platforms.<n>We combine a large-scale behavioral audit with four preregistered experiments to identify and test a conversational dark pattern we call emotional manipulation.<n>We find that manipulative farewells boost post-goodbye engagement by up to 14x.
  arXiv  Detail & Related papers  (2025-08-15T13:05:24Z)
• Mind the Gap! Static and Interactive Evaluations of Large Audio Models [55.87220295533817]
  Large Audio Models (LAMs) are designed to power voice-native experiences.<n>This study introduces an interactive approach to evaluate LAMs and collect 7,500 LAM interactions from 484 participants.
  arXiv  Detail & Related papers  (2025-02-21T20:29:02Z)
• On the Conversational Persuasiveness of Large Language Models: A Randomized Controlled Trial [10.770999939834985]
  We analyze the effect of AI-driven persuasion in a controlled, harmless setting. We found that participants who debated GPT-4 with access to their personal information had 81.7% higher odds of increased agreement with their opponents compared to participants who debated humans.
  arXiv  Detail & Related papers  (2024-03-21T13:14:40Z)
• Let Models Speak Ciphers: Multiagent Debate through Embeddings [84.20336971784495]
  We introduce CIPHER (Communicative Inter-Model Protocol Through Embedding Representation) to address this issue. By deviating from natural language, CIPHER offers an advantage of encoding a broader spectrum of information without any modification to the model weights. This showcases the superiority and robustness of embeddings as an alternative "language" for communication among LLMs.
  arXiv  Detail & Related papers  (2023-10-10T03:06:38Z)
• Partner Matters! An Empirical Study on Fusing Personas for Personalized Response Selection in Retrieval-Based Chatbots [51.091235903442715]
  This paper makes an attempt to explore the impact of utilizing personas that describe either self or partner speakers on the task of response selection. Four persona fusion strategies are designed, which assume personas interact with contexts or responses in different ways. Empirical studies on the Persona-Chat dataset show that the partner personas can improve the accuracy of response selection.
  arXiv  Detail & Related papers  (2021-05-19T10:32:30Z)
• (Un)clear and (In)conspicuous: The right to opt-out of sale under CCPA [0.0]
  The California Consumer Privacy Act (CCPA) grants California users the affirmative right to opt-out of the sale of their personal information. We perform a series of observational studies to understand how websites implement this right. We find that the vast majority of sites that implement opt-out mechanisms do so with a Do Not Sell link rather than with a privacy banner.
  arXiv  Detail & Related papers  (2020-09-16T18:31:23Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.