20.500.12556/RUL-187714 Tok zavesti in notranji monolog v besedilu, generiranem z umetno inteligenco: vprašanja subjekta, jezika in simulacije Stream of Consciousness and Internal Monologue in AI-generated Text: Questions of Subjectivity, Language, and Simulation Diplomska naloga obravnava tok zavesti in notranji monolog v literarnem besedilu, generiranem z umetno inteligenco, pri čemer raziskuje razmerje med jezikom, subjektom, zavestjo in simulacijo. Izhodišče predstavlja razlikovanje med tokom zavesti kot širšim pojavom reprezentacije notranjega življenja in notranjim monologom kot pripovedno tehniko. Teoretski del povezuje literarno-teoretske, filozofske in psihološke pristope, analitični del pa temelji na primerjavi treh besedil: Znamenja na steni Virginie Woolf, Dictée Therese Hak Kyung Cha in kratke zgodbe, generirane z velikim jezikovnim modelom ChatGPT. Analiza vzpostavi triptih treh načinov reprezentacije zavesti: pri Woolf se zavest pokaže kot razpršena, pri Cha kot vprašljiva, pri umetni inteligenci pa kot odvečna predpostavka za vzpostavitev učinka subjektivnosti. Ugotovljeno je, da z umetno inteligenco generirano besedilo reproducira diskurzivne in stilistične sledi zavesti, ne da bi bilo za njihov nastanek mogoče predpostaviti preverljiv izkustveni subjekt. Umetna inteligenca tako ne razreši vprašanja zavesti, temveč razkrije njeno povezanost z jezikovnimi učinki in odpre širši humanistični premislek o subjektu, jeziku, pomenu in literarni ustvarjalnosti. This thesis examines stream of consciousness and internal monologue in literary texts generated by artificial intelligence, exploring the relationship between language, the subject, consciousness, and simulation. It begins by distinguishing between stream of consciousness—as a broader phenomenon representing inner life—and interior monologue as a narrative technique. The theoretical section integrates literary-theoretical, philosophical, and psychological approaches, while the analytical section is based on a comparison of three texts: Virginia Woolf’s The Mark on the Wall, Theresa Hak Kyung Cha’s Dictée, and a short story generated by the large language model ChatGPT. The analysis establishes a triptych of three modes of representing consciousness: in Woolf, consciousness appears as dispersed; in Cha, as questionable; and in artificial intelligence, as a superfluous premise for creating the effect of subjectivity. It is found that AI-generated text reproduces the discursive and stylistic traces of consciousness without a verifiable experiential subject being presupposed as their source. Thus, artificial intelligence does not resolve the question of consciousness but rather reveals its connection to linguistic effects and opens up a broader humanistic reflection on the subject, language, meaning, and literary creativity. tok zavesti notranji monolog umetna inteligenca subjekt zavest jezik umetna humanistika stream of consciousness internal monologue artificial intelligence subject consciousness language artificial humanities true false false Slovenski jezik Angleški jezik Diplomsko delo/naloga 2026-09-13 07:48:32 2026-09-13 07:48:36 2026-10-02 11:09:31 0000-00-00 00:00:00 2026 0 0 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 1970-01-01 628899 293431555 73114.pdf 73114.pdf 1 0854CAC0FEAF69B0D44EBEF6B93B2F98 ea1d6361b6d27a1c82a06f4690cac7803d792d9a067213a9b20741e7308c4a59 bb3e482e-af36-11f1-8bc5-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=247419 Filozofska fakulteta 0 0 0