Oleksandr Kushnerov, a senior lecturer in the Department of Economic Cybernetics, took part in the 8th International Congress on Human-Computer Interaction, Optimization and Robotic Applications (ICHORA 2026), held on 21–23 May 2026 in Ankara, Turkey.
During the conference, our colleague presented a paper on the topic: ‘Zero-trust conceptual architecture for secure integration of large language models and autonomous AI agents’. In his work, Oleksandr analyses the challenges facing corporate security systems due to the rapid development of autonomous AI agents. Existing security paradigms, based on static access control, prove insufficient to counter the unpredictable actions of algorithmic systems. The zero-trust architecture proposed by the author is based on the use of heterogeneous graph neural networks. This allows the system to assess ‘trust’ in an AI agent in real time by analysing its behaviour and deviations from safe operating algorithms.
The results of the presentation will be published in the international scientific database IEEE Xplore.
We congratulate our colleague on a successful presentation on the international stage and wish him further scientific achievements in the field of artificial intelligence and cybersecurity!
