Artificial intelligence agents (AI agents) are computer programs or information systems that are capable of performing independent and intelligent actions while pursuing a certain goal. A critical uncharted area of AI agents relates particularly to their unethical and unwanted behavior. AI agents can be harnessed to deceive people, they can group together and be set free to carry out criminal activity, in such a way that it is not possible to stop them. The aim of the project is to identify, mitigate and prevent the negative effects of AI agents. The project investigates various manifestations of malicious artificial intelligence such as chatbots, service robots, and metaverse avatars.

2024