EU Leaders Hold Their First Summit Devoted to AI in Dublin, Hearing From Mensch and Hinton

European Council President António Costa announced the informal summit on November 3. He said the focus would be "how the EU can strengthen its competitiveness and AI sovereignty," and that "a first dedicated discussion at leaders' level is now necessary." Informal meetings of EU leaders do not take binding decisions, so the talks are meant to set the direction of the bloc's AI policy rather than settle it.

The agenda shows the tension that runs through EU technology policy. Leaders want to make it easier to build and use AI in Europe and to keep pace with the United States and China, but they also want to protect citizens and businesses from the technology's risks. Mensch will take part in a discussion on how AI can lift the European economy and how Europe can develop more of its own AI. Hinton, a former Google researcher, will join a session on protecting rights and jobs and on the security challenges AI creates.

The two speakers come from different camps. Hinton and fellow computer scientist Yoshua Bengio led the warnings about AI's dangers after OpenAI released ChatGPT in 2022. Mistral is seen as Europe's leading model developer. This year, when OpenAI's Sam Altman and Anthropic's Dario Amodei called for AI development to slow, Mistral argued the opposite and urged Europe to press ahead.

Nvidia's leadership crisis hangs over the meeting. Jensen Huang resigned as the chipmaker's chief executive on Thursday after revelations about extramarital affairs, including one with a woman accused of spying for China, and technology shares fell around the world. An EU official told AFP that Nvidia was not formally on the agenda but that nothing would stop leaders from "bringing it up". The official added that leaders would likely want to discuss any security implications of the affair.

Some governments set out their positions before the meeting. In a joint letter, Estonia and Finland said Europe needed to "establish EU-level arrangements for responding to serious AI incidents," and called for "developing advanced AI and governing it responsibly." Estonian Prime Minister Kristen Michal pointed to Europe's past dependence on outside suppliers of fossil fuels and critical minerals. "We should not make the same mistake. We should invest in new technologies," he told AFP.

The summit comes as governments debate how to regulate AI without holding back innovation. In September, major US technology companies signed a voluntary, non-binding code of conduct on AI risks, while many scientists are pressing for internationally coordinated rules.

James Green, a researcher at the Center for European Reform, said the EU needed "an all-of-the-above strategy". In his view, the bloc should "improve the business case for developing AI in Europe, strengthen its leverage to ensure access to the best models and related technologies, responsibly promote AI deployment to reap the economic benefits, and harness regulation and international partnerships to mitigate harms."