The EU AI Act [1] requires providers of AI systems to determine whether their AI system is prohibited, high-risk, limited-risk, or minimal-risk under the Act. To support this, the European Commission has published guidelines for prohibited AI [2] and high-risk systems (draft)[3] providing example use cases for each risk category. Although these guidelines are helpful, they present the use cases in a pdf, with limited searchability. This project aims to curate a linked open data resource including high-risk AI and prohibited AI use cases based on these two EU guidelines. The project builds upon existing frameworks for determining the AI Act’s high-risk AI [4] and prohibited AI systems [5] to annotate use cases and publish them as a queryable knowledge graph. Using LLM-based capabilities, the project will further develop a tool to find , within the knowledge graph, use cases similar to a free-text description of an AI system.
[1] https://eur-lex.europa.eu/eli/reg/2024/1689/oj/eng
[2] https://ai-act-service-desk.ec.europa.eu/sites/default/files/2026-01/guide-prohibited_en.pdf
[4] Golpayegani, D., Pandit, H. J., & Lewis, D. (2023, June). To be high-risk, or not to be—semantic specifications and implications of the ai act’s high-risk ai applications and harmonised standards. In Proceedings of the 2023 ACM Conference on Fairness, Accountability, and Transparency. https://dl.acm.org/doi/abs/10.1145/3593013.3594050