20.500.12556/DKUM-91632 Pozivni injekcijski napadi na velike jezikovne modele magistrsko delo Prompt injection attacks on large language models V magistrskem delu raziskujemo vlogo velikih jezikovnih modelov v vzponu generativne umetne inteligence. Predstavimo osnovne koncepte, kot so transformerji, žetoni in vektorske reprezentacije, ter opisujemo ključne prednosti, slabosti in izzive z generativnimi modeli. Posebno pozornost namenjamo izzivom varnosti, kot so pozivni injekcijski napadi. Podrobno analiziramo delovanje teh napadov, njihove vrste in predstavimo možne pristope za obrambo pred njimi. V okviru eksperimenta prikazujemo izdelavo spletne ankete, ki implementira različne jezikovne modele. S pomočjo pridobljenih podatkov iz ankete nato analiziramo občutljivost posameznih modelov na različne intenzitete injekcijskih napadov ter preučujemo njihove vplive na uporabniške dimenzije, kot so uporabnost, točnost, razumljivost in relevantnost. Na koncu ugotavljamo, kateri modeli se najbolje odzivajo na napade in predstavljajo najvarnejšo uporabo. In the master's thesis, we explore the role of large language models in the rise of generative artificial intelligence. We present fundamental concepts such as transformers, tokens, and vector representations, and describe the key advantages, disadvantages, and challenges of generative models. Special attention is given to security challenges, such as prompt injection attacks. We analyze the functioning of these attacks in detail, their types, and propose possible defense approaches. As part of the experiment, we develop an online survey application that implements various language models. Using the data collected from the survey, we analyze the sensitivity of individual models to different intensities of injection attacks and examine their impacts on user dimensions such as usability, accuracy, understandability, and relevance. Finally, we identify which models respond best to the attacks and represent the most secure usage. Generativna umetna inteligenca Generativni modeli Veliki jezikovni modeli Pozivni injekcijski napadi Inženering pozivov Generative artificial intelligence Generative models Large language models Prompt injection attacks Prompt engeneering true true false [M. Bobnar] Slovenski jezik Angleški jezik Magistrsko delo/naloga 2025-01-21 19:15:43 2025-02-06 09:19:25 2025-02-22 03:10:47 0000-00-00 00:00:00 2025 0 Maribor 0 Maribor 1 spletni vir (1 datoteka PDF (XVII, 72 str.)) 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 004.8.056(043.2) 226895619 MAG_Bobnar_Matic_2025.pdf MAG_Bobnar_Matic_2025.pdf 1 AF7B7660310E3B3E8821E2CCC74EA5FA 9620acf42fc8ddb926f527be49b7bbf7bd6dcb10b8b97119085bf7d0cfae3b0a 1b478b5d-d8e3-11ef-9853-00155d000105 https://dk.um.si/Dokument.php?lang=slv&id=188059 Fakulteta za elektrotehniko, računalništvo in informatiko 0 0 0