About the author

Mario Brcic is an Associate Professor at the Faculty of Electrical Engineering and Computing (FER), University of Zagreb, where he leads the courses Advanced Algorithms and Data Structures, the course this book grew out of, and Operations Research. He also coordinates FER’s initiatives on how a university should teach, research and work now that AI is part of all three.

His research is in AI safety and value alignment: what it takes for a system to pursue the goal its designers meant, and where that becomes provably impossible. Part of it is about specification, stating a goal precisely enough that a system can be held to it, which is the outer-alignment side of the problem, and about formal methods that check a system against what was stated. This book teaches the same discipline at the scale of a single algorithm. He co-authored Impossibility Results in AI: A Survey (ACM Computing Surveys) and the XAI 2.0 manifesto on open problems in explainable AI (Information Fusion), and he maintains the AI Safety Formalization Atlas, an open Lean library of formal AI-safety results. His doctorate, also from FER, was on coordination of large-scale projects and initiatives.

The book is also a manifestation of his research on learning. The Effortless Trap, written with Stjepan Frljic, explains why learning with AI can feel easy and teach little, and it supplies the structure these chapters follow.

Outside the university he is Managing Partner at It From Bit, an advisory practice that takes optimization and decision methods into high-stakes client work, and the author of the Cognitive Sovereignty trilogy on AI, strategy and power. Away from all of it, he runs and cycles long distances.

More at mariobrcic.com and ORCID 0000-0002-7564-6805.