On assessing trust in Artificial Intelligence systems
Abstract
The issues of trust in Artificial Intelligence (AI) systems include many aspects. Trust in AI systems is trust in their results. The results of the models used are fundamentally non-deterministic. Trust (guarantee) of the results is the stability of the model, the ability to generalize, the absence of backdoors, and many other indicators. This is where the risks of AI systems arise. Unfortunately, approaches to assessment for most of them (almost all) do not have comprehensive (final) solutions. One of the possible solutions in such a situation is to assess the very fact of using (taking into account) solutions to parry certain risks by AI system developers. We cannot assess the results of these solutions, but at least we can record attempts to solve them. What does this give? Firstly, we can assess the presence of these attempts in points, which will make it possible to compare different implementations. Secondly, parrying such risks is the best practice in the development of AI systems, accordingly, the absence of specific solutions shows developers the ways to improve their products. This is an audit of AI systems. The paper examines a European project of a questionnaire for assessing trust in AI systems, for which an adapted localized version was created, and proposed by the authors as a basis for audit systems for AI models.
Full Text:
PDF (Russian)References
Namiot D. E., Il'jushin E. A., Pilipenko O. G. Doverennye platformy iskusstvennogo intellekta //International Journal of Open Information Technologies. – 2022. – T. 10. – #. 7. – S. 119-127.
Namiot, Dmitry, Eugene Ilyushin, and Ivan Chizhov. "Ongoing academic and industrial projects dedicated to robust machine learning." International Journal of Open Information Technologies 9.10 (2021): 35-46.
Namiot, Dmitry, and Vladimir Romanov. "On improving the robustness of machine learning models." International Journal of Open Information Technologies 12.3 (2024): 88-98.
Ilyushin, Eugene, Dmitry Namiot, and Ivan Chizhov. "Attacks on machine learning systems-common problems and methods." International Journal of Open Information Technologies 10.3 (2022): 17-22.
Namiot, Dmitry. "Schemes of attacks on machine learning models." International Journal of Open Information Technologies 11.5 (2023): 68-86.
Yerlikaya, Fahri Anıl, and Şerif Bahtiyar. "Data poisoning attacks against machine learning algorithms." Expert Systems with Applications 208 (2022): 118101.
Namiot, Dmitry. "Introduction to Data Poison Attacks on Machine Learning Models." International Journal of Open Information Technologies 11.3 (2023): 58-68.
Russinovich, Mark, et al. "The Price of Intelligence: Three risks inherent in LLMs." Queue 22.6 (2024): 38-61.
Namiot, Dmitry, and Eugene Ilyushin. "On Cyber Risks of Generative Artificial Intelligence." International Journal of Open Information Technologies 12.10 (2024): 109-119.
Koshiyama, Adriano, et al. "Towards algorithm auditing: managing legal, ethical and technological risks of AI, ML and associated algorithms." Royal Society Open Science 11.5 (2024): 230859.
Namiot, D. E., and E. A. Il'jushin. "Doverennye platformy iskusstvennogo intellekta: sertifikacija i audit." International Journal of Open Information Technologies 12.1 (2024): 43-60.
Namiot, Dmitry, and Manfred Sneps-Sneppe. "On Audit and Certification of Machine Learning Systems." 2023 34th Conference of Open Innovations Association (FRUCT). IEEE, 2023.
Namiot, D., and E. Ilyushin. "On Certification of Artificial Intelligence Systems." Physics of Particles and Nuclei 55.3 (2024): 343-346.
The Assessment List for Trustworthy Artificial Intelligence https://altai.insight-centre.org/ Retrieved: Dec, 2024
Ethics Guidelines for Trustworthy AI https://digital-strategy.ec.europa.eu/en/library/ethics-guidelines-trustworthy-ai Retrieved: Dec, 2024
Radclyffe, Charles, Mafalda Ribeiro, and Robert H. Wortham. "The assessment list for trustworthy artificial intelligence: A review and recommendations." Frontiers in artificial intelligence 6 (2023): 1020592.
Fedele, Andrea, Clara Punzi, and Stefano Tramacere. "The ALTAI checklist as a tool to assess ethical and legal implications for a trustworthy AI development in education." Computer Law & Security Review 53 (2024): 105986.
Rajamäki, Jyri, et al. "ALTAI Tool for Assessing AI-Based Technologies: Lessons Learned and Recommendations from SHAPES Pilots." Healthcare. Vol. 11. No. 10. MDPI, 2023.
White Paper on Artificial Intelligence: Public consultation towards a European approach for excellence and trust https://digital-strategy.ec.europa.eu/en/library/white-paper-artificial-intelligence-public-consultation-towards-european-approach-excellence-and Retrieved: Dec, 2024
IEEE EPPC & IEEE-SA Joint Response to the European Commission AI White Paper https://globalpolicy.ieee.org/wp-content/uploads/2020/12/20005.pdf Retrieved: Dec, 2024
Namiot, Dmitry, and Eugene Ilyushin. "On the robustness and security of Artificial Intelligence systems." International Journal of Open Information Technologies 10.9 (2022): 126-134.
Namiot, Dmitry, and Eugene Ilyushin. "Data shift monitoring in machine learning models." International Journal of Open Information Technologies 10.12 (2022): 84-93.
Mohseni, Sina, et al. "Practical solutions for machine learning safety in autonomous vehicles." arXiv preprint arXiv:1912.09630 (2019).
General Data Protection Regulation GDPR https://gdpr-info.eu/ Retrieved: Jan 2025
GDPR Data Protection Impact https://gdpr.eu/data-protection-impact-assessment-template Retrieved: Jan, 2025
ISO/IEC JTC 1/SC 42 Artificial intelligence https://www.iso.org/committee/6794475.html Retrieved: Jan, 2025
THE IEEE GLOBAL INITIATIVE 2.0 ON ETHICS OF AUTONOMOUS AND INTELLIGENT SYSTEMS https://standards.ieee.org/industry-connections/ec/autonomous-systems.htm Retrieved: Jan, 2025
STANDARTY PO NAPRAVLENIJu «ISKUSSTVENNYJ INTELLEKT» https://www.rst.gov.ru/portal/gost/home/standarts/aistandarts Retrieved: Jan, 2025
Namiot, Dmitry, and Eugene Ilyushin. "Data shift monitoring in machine learning models." International Journal of Open Information Technologies 10.12 (2022): 84-93.
Bano, Muneera, Didar Zowghi, and Vincenzo Gervasi. "A vision for operationalising diversity and inclusion in AI." Proceedings of the 2nd International Workshop on Responsible AI Engineering. 2024.
Huang, Yutan, et al. "Ethical Concerns of Generative AI and Mitigation Strategies: A Systematic Mapping Study." arXiv preprint arXiv:2502.00015 (2025).
ISO/IEC 40500:2012 https://www.iso.org/standard/58625.html Retrieved: Jan, 2025
ISO/IEC Guide 71:2014 https://www.iso.org/standard/57385.html Retrieved: Jan, 2025
ISO/DIS 9241-171 https://www.iso.org/standard/39080.html#draft Retrieved: Jan, 2025
Web Accessibility Directive — Standards and harmonization https://digital-strategy.ec.europa.eu/en/policies/web-accessibility-directive-standards-and-harmonisation Retrieved: Jan, 2025
Huang, Lei, et al. "A survey on hallucination in large language models: Principles, taxonomy, challenges, and open questions." ACM Transactions on Information Systems (2024).
Namiot, Dmitry, and Eugene Ilyushin. "On Architecture of LLM agents." International Journal of Open Information Technologies 13.1 (2025): 67-74.
Namiot, D. E. Iskusstvennyj intellekt i kiberbezopasnost' / D. E. Namiot, E. A. Il'jushin, I. V. Chizhov // International Journal of Open Information Technologies. – 2022. – T. 10, # 9. – S. 135-147. – EDN DYQWEH.
Roznichnaja torgovlja v cifrovoj jekonomike / V. P. Kuprijanovskij, S. A. Sinjagov, D. E. Namiot [i dr.] // International Jou rnal of Open Information Technologies. – 2016. – T. 4, # 7. – S. 1-12. – EDN WCMIWN.
Razvitie transportno-logisticheskih otraslej Evropejskogo Sojuza: otkrytyj BIM, Internet Veshhej i kiber-fizicheskie sistemy / V. P. Kuprijanovskij, V. V. Alen'kov, A. V. Stepanenko [i dr.] // International Journal of Open Information Technologies. – 2018. – T. 6, # 2. – S. 54-100. – EDN YNIRFG.
Umnaja infrastruktura, fizicheskie i informacionnye aktivy, Smart Cities, BIM, GIS i IoT / V. P. Kuprijanovskij, V. V. Alen'kov, I. A. Sokolov [i dr.] // International Journal of Open Information Technologies. – 2017. – T. 5, # 10. – S. 55-86. – EDN ZISODV.
Refbacks
- There are currently no refbacks.
Abava Кибербезопасность ИБП для ЦОД
ISSN: 2307-8162