EU’s top AI adviser urges commissioners to use the technology to fix bloc’s economy

Por Zoya Sheftalovich — China – POLITICO

BRUSSELS — The EU should focus less on winning the race to build artificial intelligence and more on how to deploy it to turn around the bloc’s sluggish productivity, the European Commission’s chief AI adviser told commissioners.

The technology is Europe’s best chance to accelerate growth in a range of sectors — including health, agriculture, transport, defense and manufacturing — Jim Hagemann Snabe, the EU’s new special envoy for industrial AI, told the bloc’s 27 commissioners at a Sept. 4 seminar, according to two Commission officials. They were granted anonymity to discuss the private briefing.

The EU is trying to capture the economic gains of rapidly advancing AI even as it grapples with the technology’s risks to jobs, security and society. Europe’s leaders should not worry too much about leading the development of frontier AI models, but should “be the ones who draw the greatest value from it” while managing its dangers, Commission President Ursula von der Leyen said in her State of the Union speech last week.

The Commission brought Snabe in as special envoy in July to advise von der Leyen and Executive Vice President Henna Virkkunen on industrial AI and shape the EU executive’s strategy on the technology. His mandate spans the infrastructure and technologies needed to deploy AI — from data centers, computing capacity and chips to models and applications across industry.

A week before Snabe addressed the commissioners, a Commission AI expert delivered a separate seminar for the commissioners’ heads of cabinet. The Commission declined to name the expert as they are employed in a technical, not political capacity.

The message at both gatherings was to shift the debate from whether Europe can build the world’s leading AI models to how it can use the tech to drive its economy.

The seminars offered an early look at how that work is shaping the Commission’s approach.

Shift to adoption

The U.S. and China may lead in developing models and building computing capacity, the AI expert told the cabinet chiefs at their Aug. 28 seminar, but Europe can still benefit by applying AI in health care, manufacturing and public services, according to slides from the meeting seen by POLITICO. The presentation called for Europe to “invert the logic” of the debate.

That means identifying the most important uses for AI, pinpointing what is blocking them and taking targeted steps to remove those barriers, the expert said. Examples included rules to make it easier to test autonomous vehicles before mass production, secure data spaces for personalized medical treatment and a digital “farmer co-pilot” offering advice based on live farm and remote data.

The AI expert left the cabinet chiefs with what they called “food for thought”: What could AI make possible in their portfolios that was previously too expensive or impossible? Which assumptions might no longer hold if cognitive work becomes dramatically cheaper? And how could policy promote positive outcomes while guarding against harmful ones?

AI could also help Europe accelerate the economic changes set out in Mario Draghi’s influential 2024 competitiveness report by speeding up experimentation, selection, scaling and diffusion of new ideas, the presentation said, with the expert pointing to slowing labor productivity, an aging population, weaker trade and the need to renew infrastructure as headwinds to growth. The goal is to extend the Draghi agenda from the factors of competitiveness to the speed of economic evolution, they said.

But useful applications will take time to develop. That means it must become easier for companies to experiment, grow and scale successful ideas, while ensuring Europe retains a share of the value created, rather than becoming dependent on a handful of outside providers for models, chips or computing power, both Snabe and the AI expert said.

The seminars are feeding into a broader Commission effort to turn that ambition into a plan. The Commission will until December undertake “deep work” mapping out how to use AI across different sectors. Snabe and his team will conduct sector analyses, consider pilot projects and examine the AI “tech stack” — models, chips and computing capacity — before drafting a policy package toward the end of the year. A final report and strategy should take shape in early 2027, and the team is due to examine the societal implications of the policies between January and March.

Commission spokesperson Thomas Regnier said Snabe’s Sept. 4 seminar “focused on how Europe can accelerate the uptake of industrial AI.” He added that the special envoy “will deliver an evidence-based and forward-looking report on Industrial AI.”

The Commission’s evolving approach to AI comes as critics question whether it is keeping pace with the technology’s risks. Axel Voss, a German conservative lawmaker and von der Leyen ally, said the Commission had failed to adapt to rapid developments and was not enforcing the AI Act — the EU’s law governing the risks of the technology — effectively enough. He pointed to recent hacking incidents involving AI agents and called for rules on liability when AI causes harm.

Cardinal Michael Czerny, who chairs the Vatican’s commission on AI, urged European politicians and bishops to respond more clearly to warnings that frontier models could escape human control. Czerny said he had not yet used AI himself, but felt it was his “moral duty” to gain personal experience with the technology.

Fonte: China – POLITICO