AI i samhällets tjänst – Balansgång mellan innovation och risk inom samhällsviktiga verksamheter

Sammanfattning

This study examines various risks associated with the use of artificial intelligence (AI), which have become increasingly relevant as the technology rapidly develops and integrates into society. The risks have been categorized into six main areas: technological, data- and analytical, informational and communicational, economic, social, ethical and legal and regulatory. These areas include challenges such as technological complexity, the spread of false information, economic losses, social stress, discrimination, and difficulties navigating current legislation. What is also highlighted is the upcoming EU AI Act, which aims to regulate AI systems by categorizing them based on their level of risk. The results from literature review, surveys, and interviews with 19 respondents knowledgeable in AI from critical societal operations show that the risks are currently low but are expected to increase. This is partly due to the increased use of advanced, generative AI technologies that are more opaque and complex. The study also highlights similarities and differences in how the private and public sectors manage these challenges. Particularly in public operations, and to some extent in some private organizations, data often subject to stricter legal requirements is handled. This sensitive data management imposes high demands and can risk limiting how AI can be utilized, potentially creating obstacles for operations where the need for efficiency is great. Finally, the challenge that organizations must face is emphasized as they need to balance the utilization of AI for its many benefits while also managing and minimizing the risks that the technology entails. This balancing act is a complex and crucial challenge in the development and implementation of AI technology.