Lessons Learned from Co-governance Approaches – Developing Effective AI Policy in Europe

Abstract

Advancements in artificial intelligence (AI) are moving faster than the State’s ability to fully govern it, resulting in a need for innovative approaches that also involve non-state actors, or “co-governance” mechanisms. But the question remains as to exactly how co-governance mechanisms can be incorporated into AI governance in the EU. While governing the use of AI presents certain challenges to traditional modes of governance, these challenges are not unique to AI. This chapter provides an in-depth case study of co-governance mechanisms used in relation to services and supply chains that have similar governance challenges to AI in order to map a spectrum of co-governance mechanisms that may be useful for governing AI. After analyzing the applicability of various co-governance mechanisms to the AI governing space, the chapter offers recommendations for how and where these approaches may be most worthwhile. Effective AI governance will require a combination of approaches. Key to success will be State support, cooperation or buy-in in order to maintain legitimacy, while capitalizing upon non-state capacity and distributed implementation. Specifically, first, States, educational institutions and companies should make a major push to integrate responsible AI practices in education curriculums, setting the foundation for effective accreditation and professionalization mechanisms that employers recognize and value. Second, for lower-risk applications, EU Governments should support or compliment multi-stakeholder initiatives to assess and label products. Finally, for higher-risk AI systems, legislative points of control that include the use of co-governance mechanisms for implementation should be put in place.