Truth, persuasion, and AI design for human users | IASEAI '26

How do AI systems shape what we believe—and who we trust? This breakout session from IASEAI 2026 explores how large language models influence human judgment, from persuasion and debate to information sourcing and societal impact. Speakers examine how AI systems form and express beliefs, how they prioritize information sources, and how their outputs may subtly shape user understanding in high-stakes contexts. Across two segments, the session highlights emerging research on persuasion dynamics, historical accuracy, latent biases, and system-level approaches to influence—alongside the implications for governance, accountability, and human-centered AI design. María Victoria Carro — AI Debaters are More Persuasive when Arguing in Alignment with Their Own Beliefs Francesco Ortu — Preserving Historical Truth: Detecting Historical Revisionism by Large Language Models Soumi Das — In Agents We Trust, but Who Do Agents Trust? Latent Source Preferences Steer LLM Generations Maximilian Kroner Dale — DeliberationBench: A Normative Benchmark for Assessing the Influence of Large Language Models on Users’ Views Arthur Vogels — In-Distribution Steering: Balancing Control and Coherence in Language Model Generation Zuzanna Wojciak — TRIED: Truly Innovative and Effective AI Detection Benchmark Kaylee Williams — Deepfake Financial Fraud: The Global Regulation of AI-Driven Scams Anna Gausen — Disclosure by Design: Identity Transparency as a Behavioural Property of Conversational AI Models 📍 Recorded at UNESCO Headquarters, Paris The International Association for Safe and Ethical AI (IASEAI) is a global professional association bringing together researchers, policymakers, and practitioners working to ensure that advanced AI systems operate safely and ethically for the benefit of humanity. Learn more: https://www.iaseai.org #AI #AISafety #AIGovernance #AIpolicy #IASEAI #ArtificialIntelligence #ResponsibleAI