On 19 May 2026, at the Ministry of Digital Governance, GR digiGOV-innoHUB hosted Connect: Proof of Impact. The title reflected the ambition of the day: to move the conversation beyond promises and onto results. What has Greece actually built around artificial intelligence? What pieces are already in place? And what will determine whether these investments translate into real impact?

What made the discussion particularly interesting was that it brought together representatives from the entire AI value chain, from infrastructure and data to language technologies, public policy and deployment.

Several speakers highlighted how the foundations are now being put in place. Among them Panos Louridas (GRNET), Vassilios Peristeras (International Hellenic University), Vassilis Katsouros (Athena Research Center), and Petros Stefaneas (NTUA / GlossAPI-GFOSS). They argued that national initiatives such as the DAEDALUS supercomputer and PHAROS are creating capabilities that did not exist a few years ago. At the same time, progress on the European AI Act is helping establish a framework for trustworthy AI deployment. Moving from “test before invest” to “deploy with impact” is where the day pointed next, in the session on AIPOWERGREECE with Special Secretary  for Artificial Intelligence and Data Governance Vassilis Karkatzounis and Periklis Terlixidis (NCSR Demokritos). The discussions concluded with remarks from Dimitris Papastergiou, Minister of Digital Governance  who referred to the upgraded data.gov.gr platform and noted that the real challenge lies in building a culture around open data.

Taken together, these interventions pointed to the same conclusion: Greece is gradually assembling the building blocks required for a national AI ecosystem. The question is how these pieces can be connected and transformed into real-world impact.

The discussion also exposed a series of persistent obstacles. Public services remain largely organised around documents rather than data. Valuable information is often fragmented across institutions, difficult to access, and not readily usable for AI systems. Interoperability standards exist but are unevenly adopted. Language technologies for Greek continue to advance, but questions remain about long-term sustainability, funding and deployment at scale.

For Vangelis Karkaletsis, President of NCSR Demokritos, these challenges point to a broader issue: the need to move from viewing AI as a technological achievement to treating it as public infrastructure.

Drawing on the foresight study Generative AI Greece 2030, he argued that foresight is not about predicting a single future. It is about preparing for multiple possible futures and identifying the choices that make desirable outcomes more likely.

Among the scenarios explored in the study, techno-social acceleration stands out as both the most desirable and increasingly plausible trajectory for Greece. The country is not there yet, but important building blocks are beginning to align. Advanced computing infrastructure, national AI services, evolving regulatory frameworks and a growing research base are gradually forming an ecosystem capable of supporting large-scale transformation.

Two developments make this moment different.

The first is cultural. A new generation of researchers is increasingly focused not only on producing knowledge but also on understanding how research can become a service, a product or a solution to a world problem. This shift brings scientific excellence closer to societal impact.

The second is structural. Through initiatives such as PHAROS, Demokritos is no longer operating solely as a research institution. As technical coordinator, it occupies a position where scientific expertise, computational infrastructure, public-sector needs and business capabilities can meet. The challenge now is to turn these connections into practical outcomes.

At the same time, trustworthy AI should not be viewed as a constraint on innovation. Particularly in the public sector, trust is a prerequisite for adoption. Transparency, accountability and reliability are essential if AI systems are to become part of everyday public services.

Yet infrastructure alone will not drive progress. The existence of computing capacity, research excellence and talent does not automatically lead to adoption. In that sense, the next phase for Greece is more about implementation. The challenge is to connect infrastructure, data, expertise and public-sector demand in ways that produce measurable results. Pilot projects, operational deployments and closer collaboration between government, research organisations and industry will determine whether the country’s growing AI capacity translates into public value.

The broader ambition is that AI’s potentials should not become available only to a small number of organisations or experts. Its benefits must reach citizens through better services, and more efficient and responsive public administration.

Many of the foundations are now in place. The next step is to transform capacity into implementation and research excellence into outcomes that people experience in their daily lives.