Μake us preferred on Google For a country that is rarely mentioned when the world talks about artificial intelligence, Greece has been turning up in some interesting places.
On September 28, Prime Minister Kyriakos Mitsotakis endorsed an international call for greater control of frontier AI models, joining leaders from more than 30 countries. The statement calls for mandatory pre-deployment testing, independent evaluation, common standards for reporting serious AI incidents and, more ambitiously, exploration of an international institution capable of setting standards and enabling verification. That might sound like another declaration in an already crowded field of AI principles. It is not insignificant.
The statement goes directly to one of the central questions now emerging in global AI governance: how do governments retain meaningful oversight of systems whose capabilities are advancing faster than the institutions meant to govern them?
Greece is beginning to have something to say about that.
The timing matters. At the United Nations General Assembly this month, Mitsotakis argued that voluntary self-regulation by the technology industry is insufficient and called for frontier AI companies to face licensing requirements based on transparency and safety. He explicitly compared the challenge to the regulation of nuclear technology, while acknowledging that the United States, where most frontier models are developed, has a particularly important responsibility.
Foreign Minister George Gerapetritis went further in New York, co-chairing a Greek-Indian UN event on AI, emerging technologies and international peace and security. He argued that AI governance is no longer simply a technology question but one concerning international security, responsibility and international law. Greece, he said, intends to continue that conversation within the United Nations.
This is not happening in isolation. Greece has already participated in international discussions on AI and security, including an April 2025 UN Security Council meeting it convened with France and South Korea. It has developed a national AI and data-governance structure, participates in European AI policymaking and has invested in AI infrastructure.
And there is another signal that should not be overlooked. In June, Greece joined the Pax Silica initiative on AI and strategic technology supply chains. Greece also faces an unusual diplomatic opening: its role in the United Nations and its forthcoming presidency of the Council of the European Union create platforms from which it could move beyond implementing other people’s rules and help shape them.
Taken together, these are more than isolated gestures but, at the same time, they are not yet an AI foreign policy.
Global AI governance is becoming a contest not only over regulation but over institutional power. Who gets to define what constitutes a safe frontier model? Who evaluates it? Who verifies claims made by developers? Who establishes the standards that governments eventually adopt? Who has the technical capacity to produce evidence that others trust?
These questions are increasingly being negotiated internationally. The recent call on frontier AI makes this explicit: it asks governments to build shared scientific capacity and trusted evaluation and to consider an institution that could provide verification.
This is an area where Greece could have something distinctive to contribute. It does not need to pretend that it can compete with the United States or China in frontier-model development nor should it try to manufacture a claim to technological power it does not possess. Middle powers rarely gain influence by pretending to be great powers. They gain it by becoming useful.
Greece could become useful in the emerging architecture of AI governance by concentrating on areas where it can build genuine credibility: AI evaluation and safety, international standards, maritime and critical-infrastructure security, the application of AI in peace and security, and the institutional links between European regulation and global governance. The point is to develop expertise that other governments actually need.
That could mean building a serious national capacity for evaluating advanced AI systems. It could mean using Greece’s position on the UN Security Council to push the conversation from abstract principles toward verification, incident reporting and accountability. It could mean using the 2027 EU Council presidency to connect European AI regulation with the wider international debate rather than treating the two as separate tracks. And it could mean making technology diplomacy a genuine part of Greek foreign policy.
There is already evidence that Athens understands the direction of travel. The question is whether the government sees these individual initiatives as parts of the same strategy. Because the danger for Greece is not irrelevance. It is fragmentation.
A country can sign an international declaration, host a conference, participate in a UN discussion and implement European legislation without ever developing a coherent position on what kind of global AI order it wants. That would be a missed opportunity.
The global debate is moving quickly. The question is no longer whether AI will require international governance. The question is who will shape the institutions, standards and verification mechanisms that eventually emerge. Greece has started showing up. Now it needs to decide whether it wants simply to be in the room or whether it wants to have something useful to say once it gets there.
Dr. Konstantinos Komaitis is a resident senior fellow with the Democracy and Technology Initiative at the Atlantic Council, leading the work on digital governance and democracy.
