20.500.12556/RUL-174431 Vprašanja odgovornosti v postopkih samo-triažiranja s pomočjo umetne inteligence (magistrsko diplomsko delo) Liability issues in self-triage procedures using artificial intelligence Magistrska naloga obravnava pravne izzive, ki jih prinaša uporaba umetne inteligence v postopkih (samo)triaže v medicini. Tehnološki napredek odpira možnosti za razbremenitev zdravstvenega sistema, hkrati pa izpostavlja nevarnosti, povezane z nezanesljivostjo, nerazložljivostjo odločitev in pristranskostjo sistemov. Pravna znanost se pri tem sooča z vprašanjem, kako obstoječi odškodninski režimi naslavljajo tveganja, ki jih ustvarja takšna tehnologija. V raziskavi je analiziran evropski regulativni okvir, zlasti Akt o umetni inteligenci in prenovljena Direktiva o odgovornosti za proizvode, ter slovenska ureditev medicinske odgovornosti. Ugotovljeno je, da evropski pristop poudarja predvsem obvladovanje tveganj in določitev minimalnih standardov varnosti, ne pa harmonizacije materialnih pravil civilnega prava. To pomeni, da se odgovornost za napake sistemov umetne inteligence v medicini še naprej presoja predvsem po pravilih krivdne odgovornosti, ki pa v praksi pogosto otežujejo dokazovanje. Možnosti objektivne odgovornosti ostajajo omejene in niso prilagojene posebnostim samoučečih se sistemov. This master’s thesis examines the legal challenges posed by the use of artificial intelligence in (self-)triage processes in healthcare. Technological advances offer opportunities to ease the burden on healthcare systems, but at the same time raise concerns about the unreliability, opacity, and bias of AI systems. Legal scholarship is thus faced with the question of how existing liability regimes address the risks created by such technologies. The research analyses the European regulatory framework, particularly the Artificial Intelligence Act and the revised Product Liability Directive, as well as the Slovenian rules on medical liability. It finds that the European approach focuses primarily on risk management and setting minimum safety standards, rather than harmonising substantive civil-law rules. As a result, liability for errors of AI systems in medicine is still largely assessed under fault-based principles, which in practice often make proof difficult. The possibilities for strict (objective) liability remain limited and are not adapted to the specific features of self-learning systems. umetna inteligenca – samotriaža – triaža – odškodninska odgovornost – Akt o umetni inteligenci – Direktiva o odgovornosti za proizvode artificial intelligence – self-triage – triage – liability – Artificial Intelligence Act – Product Liability Directive true false false D. Ljube Slovenski jezik Angleški jezik Magistrsko delo/naloga 2025-10-02 09:00:07 2025-10-02 09:00:11 2025-11-22 04:07:37 0000-00-00 00:00:00 2025 0 Ljubljana 0 Ljubljana [96] str. 0000-00-00 Zaloznikova Objavljeno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 347:004 131437 257914115 25851.pdf 25851.pdf 1 4A7653409D0AB5C1FCBCEC8060CE3CA7 e84d3ea5201e7cd53bb3cacf0c374bbdf8c212d2b6c4b5550770eb852fec147b 3ae2445b-9f5d-11f0-9328-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=219235 Pravna fakulteta 0 0 0