João Donato gave a presentation entitled “Benchmarking LLM robustness against prompt-based adversarial attacks”, as part of our ongoing collaboration with the MINSAIT/INDRA group. João recently completed his M.Sc. with us, and INDRA, who co-sponsored his work, wishes to highlight this collaboration to a broader national audience.
The presentation was be followed by a roundtable discussion with Helena Peixoto (CIO at Plural+Udifar), Ana Ferreira (RISE/FMUP), and António Ribeiro (Head of Cybersecurity at MINSAIT/INDRA) on the challenges of using and integrating AI agents in operational contexts.
Software and Systems Engineering (SSE)
Centre for Informatics and Systems (CISUC)
