Malicious AI swarms pose a direct threat to information integrity, institutional trust, and the reliability of AI training data. These coordinated networks of autonomous agents, built on large language models, have moved from research concern to documented reality: influence campaigns using AI-generated content targeted elections in Taiwan, India, and the United States in 2024.
CXOTalk episode 915 explores the impact of AI swarms with Daniel Thilo Schroeder, Research Scientist at SINTEF Digital, and Jonas R. Kunst, Professor of Communication at BI Norwegian Business School.
Schroeder and Kunst co-led a 22-author study published in Science in January 2026 that maps these threats and outlines specific defenses. Their framework explicitly distinguishes between what is empirically established and what remains uncertain, strengthening its value for decision-makers.
What we will cover:
- How AI swarms differ structurally from earlier bot networks, and why previous influence operations are an unreliable baseline for assessing current risk
- The "LLM Grooming" threat: how adversaries flood the web with fabricated content designed to corrupt AI training data at the next model retraining cycle
- How platform business models create misaligned incentives, where inauthentic accounts inflate engagement metrics that drive revenue, sustaining the threat and complicating governance
- Why disrupting the commercial market for influence operations is more effective than regulatory mandates alone, and what platform providers and enterprise technology buyers must do in response to these threats
- What defensible AI governance looks like in practice: detection mandates, provenance standards, and the specific risks of government-controlled counter-messaging tools
- How synthetic consensus exploits the psychology of social proof, making executives, employees, and citizens vulnerable even when they believe they are skeptical
Episode Participants
Daniel Thilo Schroeder is a research scientist working at the intersection of AI, computational social science, and digital risk. His research examines how emerging technologies reshape information ecosystems, democratic resilience, and societal security, with a particular focus on coordinated AI-mediated influence operations and multi-agent dynamics in online environments. He combines large-scale social media and behavioral data analysis with simulation-based approaches to study how coordinated campaigns spread, how influence systems evolve, and how institutions can respond.
Jonas R. Kunst is Professor of Communication at BI Norwegian Business School and Professor of Cultural and Community Psychology at the University of Oslo. His research examines misinformation and conspiracy theories, violent extremism, and the psychological implications of artificial intelligence. He previously was a Fulbright scholar at Harvard and a postdoctoral fellow at Yale. His work has been published in Science, Nature Communication, Nature Human Behavior, PNAS, Psychological Science, and other leading journals.
Michael Krigsman is a globally recognized analyst, strategic advisor, and industry commentator known for his deep business transformation, innovation, and leadership expertise. He has presented at industry events worldwide and written extensively on the reasons for IT failures. His work has been referenced in the media over 1,000 times and in more than 50 books and journal articles; his commentary on technology trends and business strategy reaches a global audience.
0 Commentaires