2026-09-21T18:27:36Z https://www.ideals.illinois.edu/oai-pmh

oai:www.ideals.illinois.edu:2142/124375 2024-09-16 col_2142_5131 col_2142_10761 com_2142_5130 com_2142_10755 com_2142_234

Forsyth, David Forsyth, David Li, Bo Lazebnik, Svetlana Krueger, David 2024-05 application/pdf en text Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2024-09-16 without embargo terms The student, Mantas Mazeika, accepted the attached license on 2024-04-23 at 09:49. The student, Mantas Mazeika, submitted this Dissertation for approval on 2024-04-23 at 09:57. This Dissertation was approved for publication on 2024-04-24 at 15:16. DSpace SAF Submission Ingestion Package generated from Vireo submission #20566 on 2024-09-16 at 00:35:51 Toward managing catastrophic AI risks Mazeika, Mantas 2024-04-24 Ai Safety Ai Risk Robustness Red Teaming Neural Trojans Trojan Detection Alignment Model Stealing Artificial intelligence (AI) has rapidly improved over the past decade, leading to widespread adoption of AI systems and demonstrating the potential for AI to greatly benefit society. However, as with any powerful new technology, AI introduces risks that must be managed to fully realize these benefits. Recent breakthroughs in the generality of AI systems have drawn increased attention to AI risks, including those of a potentially catastrophic nature. To help manage these anticipated risks, we take a defense in depth approach, combining different areas of AI safety research to address different aspects of AI risk. We present research on making AI systems more robust to adversarial influence, monitoring AIs for hidden behavior and trojans, enabling AIs to understand and adhere to human values, and finally addressing systemic problems to enable increased transparency. Text eng https://hdl.handle.net/2142/124375 Copyright 2024 Mantas Mazeika Ph.D. Dissertation Computer Science University of Illinois at Urbana-Champaign Computer Science