ai
Toward Meaningful Transparency for AI Chatbots: Disclosing Persuasive Intent Reduces Persuasion
- Source
- arXiv — Computers and Society
- Published
- Last verified
- 13 Aug 2026
- Confidence
- High
- Evidence
- Original document retained
- Reading time
- 1 min
- Country
- International
- Relevant to
- Policy & Regulation, Research & Evidence, Technology & Data
Executive summary
What happened, and why should leadership care?
Recent research from arXiv investigates the effectiveness of disclosing AI involvement and persuasive intent in chatbot interactions on user persuasion. An experiment with 1,500 UK adults revealed that while a simple AI disclosure did not significantly alter a chatbot's persuasive impact, explicitly stating the chatbot's persuasive intent, alongside AI disclosure, effectively reduced its persuasive influence. This suggests that the nature and specificity of transparency mechanisms are critical in managing user responses to AI-driven communication.
Why this matters
Why is this strategically important?
This research provides critical insights for organizations deploying AI chatbots, particularly in communication and engagement roles. Understanding how different transparency disclosures affect user receptiveness to AI persuasion is vital for maintaining trust, ensuring ethical AI deployment, and navigating evolving regulatory expectations around AI content provenance and intent.
Key insights
What should be noted from the evidence?
- AI chatbots possess a significant persuasive capability, demonstrated by a 12.6-point shift in attitudes among participants.
- A general disclosure indicating interaction with an AI system (T1) did not reduce the chatbot's persuasive appeal.
- A specific disclosure combining AI involvement with an explicit statement of persuasive intent and instructions (T2) effectively reduced the chatbot's persuasive impact.
- The study highlights that merely identifying content as AI-generated may be insufficient to counteract persuasive attempts; transparency regarding intent is crucial.
Evidence and confidence
How far can this assessment be trusted?
High confidence. Named institution, original document retained and analysis corroborated.
Analysis is prepared editorially by Aziz Shuaib Ausi. The original publication remains the authoritative record, and executive judgement remains entirely human.
Source
Where does this originate?
Reported by arXiv — Computers and Society · International. This briefing summarises the publication for executive use; the document itself is not reproduced here.
Read the original publication