
Presenter: Federico Cerutti
Monday, 13 July, 2:30 PM
Location: Sala Conferenze “Luigi Ciminiera”, DAUIN, Politecnico di Torino
ABSTRACT
Ransomware extortion is both a technical attack and a negotiation in which attackers seek payment and victims try to limit loss, gain time, and test the threat’s credibility. In this talk, the speaker will discuss publicly leaked ransomware negotiations as argumentative dialogues and analyse them through dialogue types, dialogue moves, and argumentation schemes. He will describe a fully automated, two-stage, LLM-based adjudicated annotation workflow for hostile conversational data. The analysis shows that ransomware negotiation is not mere coercion: negotiation dialogue and practical reasoning dominate, while value appeals, causal forecasts, evidential claims, and procedural manoeuvres also shape the exchange. Ransomware negotiation analysis with computational argumentation requires methods designed for compressed, strategic, and coercive dialogue.
BIOGRAPHY
Federico Cerutti is Full Professor at the University of Brescia and leads the Brescia branch of the CINI National Cybersecurity Lab. His work focuses on AI, cybersecurity, and reasoning under uncertainty, especially on systems that help people understand not just what a model recommends, but how reliable that recommendation is and what evidence supports it. He is also affiliated with Cardiff University, the University of Southampton, and Imperial College London.