As rotating president of the Security Council in October, Greece is planning to draw attention to a topic close to Athens’ heart: Artificial intelligence and maritime security. Latvia, which holds the Council presidency of the 15-member body in November, is also expected to highlight artificial intelligence in a broad thematic discussion on emerging technologies.

Together, the two countries also intend to announce the creation of an informal working group in the Council on technology, peace and security during the third week of October, Greece’s UN mission told PassBlue.

During Latvia’s presidency in November, it will invite fellow Council members to pledge support to the new working group’s ambitions.

The initiative would connect various technological concerns to specific Council agenda items, giving members and other countries a consistent forum for addressing these issues. Greece described the working group to PassBlue as a contribution intended to outlast its Council term, which ends on Dec. 31.

So far, the European countries in the Council have been using their monthly rotating presidencies to not only organize briefings and informal meetings on AI but to also bring in experts and civil society on the matter and to build coalitions with other member states to give AI political prominence despite disagreements on possible global regulations and the related AI  rivalry between China and the US.

In September, France, a permanent member and rotating president of the Council that month, convened a high-level open debate on the seeming advantages and dangers of frontier technology, with the leaders of Anthropic and OpenAI, the leading American AI companies, participating in the session.

In the Council, the European countries will try to rally their fellow members behind discussions on frontier technology as political tensions harden on the role the Council should be playing in shaping the future of the technology to maintain international peace and security.

“One of the interesting things about elected members is that they can have more influence over the Council’s agenda than their formal status might suggest,” Niels Nagelhus Schia, the head of Research Group for Security and Defense at the Norwegian Institute of International Affairs told PassBlue. “They cannot determine the Council’s priorities on their own, but they can act as agenda-setters and coalition-builders.” (Besides Greece and Latvia, the other current elected members are Bahrain, Colombia, the Democratic Republic of the Congo, Denmark, Liberia, Pakistan, Panama and Somalia.)

UK speaks up first 

Long before it became an active conversation in the Security Council, concern from the industry itself about potential risks of AI set off growing angst among world leaders during the May 2023 Group of 7 leaders. At the group’s summit in Hiroshima at the time, the participating governments committed to advancing international discussions on AI governance.

Shortly after the meeting, Rishi Sunak, then the British prime minister, met the chief executives of OpenAI, Google DeepMind and Anthropic to talk about the technology’s existential threats and international safety measures, among other matters.

The topic was further pursued by Britain in its presidency of the Security Council in July 2023. In what became the body’s first meeting on AI, the UK pushed for a wide diplomatic role on AI safety. This marked Europe’s first attempt to bring the continent’s concerns about the frontline technology into the Council.

“As humanity stands on the precipice of this gigantic technological leap forward, we want to encourage a multilateral approach to managing both the huge opportunities and the risks that artificial intelligence holds for all of us,” Barbara Woodward, Britain’s envoy to the UN at the time, said on July 3, 2023.

Since the launch of ChatGPT in 2022, the Council has held seven meetings on AI, and four of them have been organized by a European country either alone or with other Council members. Greece’s meeting on AI and maritime security, scheduled for Oct. 29, will raise the total count of AI meetings to eight.

Greece’s debate will explore the implications of AI and other emerging technologies for global shipping and freedom of navigation, an especially timely topic given the monthslong crisis in the Strait of Hormuz and the Bab al-Mandab Strait in the Mideast. The focus of Greece will draw on its longtime maritime expertise and its previous Council engagement on maritime security, also related to freedom of navigation.

Inevitable disagreements

Sanita Pavluta-Deslandes, Latvia’s permanent representative to the UN, told PassBlue that the consecutive Council presidencies of Denmark, France, Greece and Latvia have been providing the European countries the chance to sustain and build on one another’s work on AI.

“We cooperate and coordinate closely throughout our terms, and especially when preparing for our respective presidencies,” Pavluta-Deslandes said. “This provides an opportunity to ensure that emerging technologies and AI are approached with continuity, complementarity and a longer-term perspective.”

However, the national priorities of each country will shape which aspects of AI receive attention in the Council. Besides Greece’s interest in marine security, France highlighted AI safety, particularly for children; Latvia plans a wider conversation on emerging technologies.

While these discussions can make AI’s implications more focused, said Rodrigo Saad, a policy analyst at the research group Security Council Report, but he cautioned that disagreement might arise when national interests conflict.

The Council lacks both a mandated regular briefing on AI and corresponding reports by the UN secretary-general on how the technology affects the body’s agenda items. Wider disagreements, such as whether debates on AI governance belong in the Council as well as opposition from a permanent member could constrain European ambitions to push for consensus on formal results like a resolution or a lesser product, such as a statement, Saad said. European countries have pushed the hardest for common rules, human-rights safeguards and human control on AI.

Similarly, Duncan Cass-Beggs, the executive director of the Global AI Risks Initiative at the Toronto-based Center for International Governance Initiative (CIGI), said that the Europeans’ plans to bring AI more into Council discussions are valuable, but the attention must translate into safeguards against the gravest risks. These include the loss of human control over powerful autonomous systems.

Secretary-General António Guterres warned in an Oct. 8 report that advances in neurotechnology combined with artificial intelligence could threaten freedom of thought, privacy, protection against self-incrimination and the right to a fair trial.

Cass-Beggs said that Europe would need to extend its Council engagement to make cooperation politically and practically possible between Washington and Beijing. Moreover, he added that European countries’ domestic regulations alone cannot prevent dangerous developments at the leading AI laboratories.

“The most important thing for the safety and security of [European] citizens’ right now is actually that there’s an agreement between the United States and China,” Cass-Beggs told PassBlue in an interview on Oct. 6. “We need to move in that direction and not in the direction of building uncontrollable, autonomous, superintelligent agents.”

Council members broadly agree that AI offers benefits alongside serious risks, but they disagree on who sets the rules, how restrictive they should be and whether safety controls could reinforce unequal access to technology.

The solution, Saad of the Security Council Report said, is establishing an informal expert group on technology, peace and security similar to what Greece and Latvia are putting together. Council members’ specialists could meet informally before mandate renewal negotiations on agenda items begin.

These informal meetings could include briefings from relevant UN bodies and UN field operations about potential technological risks — such as AI-driven disinformation — affecting the many countries on the Council’s agenda

“The Informal Expert Group will consider both the challenges and opportunities arising from technological developments, with a particular focus on their implications for the Council’s work,” Latvia’s ambassador told PassBlue. “It is intended to complement, rather than duplicate, discussions taking place in other international formats.”