Two years after bringing the AI Act to the forefront, Europe’s focus is now shifting from framework to implementation. Vangelis Karkaletsis, Director and Chairman of the Board of NCSR Demokritos, and Alexandros Nousias, Research Associate on AI Law, Governance & Ethics at NCSR Demokritos, contributed to the discussion “AI Governance in Practice: Implementing the AI Act and enabling innovation in the EU”, held in Brussels on March 19th. The event, organised by the Special Secretariat for AI and Data Governance of the Hellenic Ministry of Digital Governance and its Special Secretary Vasilis Karkatzounis, together with the Permanent Representation of Greece to the EU, brought together representatives of European institutions, national authorities and industry to explore what that shift means in practice.
AI governance is not a policy exercise in isolation. It has direct implications for how research is conducted, how systems are developed, and how trust is built across the ecosystem. Key themes on the table were coordination between national authorities and EU-level actors, the practical challenges of operationalising the AI Act, balancing innovation with effective safeguards, and emerging issues such as deepfakes and trust in the information space.
Vangelis Karkaletsis highlighted the strategic importance of aligning AI governance with Europe’s innovation agenda and the need to strengthen institutional capacity for the effective implementation of the AI Act. As he noted: “Strengthening the adoption of AI by SMEs, industry, and the public sector is a key national priority within our national AI strategy. To that end, Greece is actively participating in European initiatives such as the European Digital Innovation Hubs (EDIH), as well as AI Factories, including the Greek AI Factory ‘Pharos’. Today’s discussion is therefore particularly significant, as the challenge is no longer only to develop AI, but to develop trustworthy AI.”
Alexandros Nousias moderated the panel “Enabling Innovation within Governance”, where the discussion explored the concept of trustworthy AI and its practical application in the field, including in the context of the Pharos AI Factory. The panel framed trustworthy AI not merely as a compliance objective, but as a system of quality governance with values engineered at its core, enabling innovation to scale with trust, accountability, and societal alignment.
The event brought together participants from Cyprus, Ireland, Italy, Lithuania, and Austria, as well as the EU AI Office.