By Jeffrey Dastin and Elizabeth Pineau
PARIS (Reuters) – All eyes are on the French capital next week to see if U.S. President Donald Trump’s administration can find common ground with China and nearly 100 other nations on the safe development of artificial intelligence.
About a year after world powers reckoned with the dangers of AI in England’s Bletchley Park, a wider array of countries are gathering in Paris to discuss putting the technology to work.
France, eager to promote its national industry, is hosting the AI Action Summit alongside India on Feb. 10 and 11, with a focus on areas where Europe’s second-largest economy has an advantage: freely available or “open-source” systems, and clean energy to power data centers.
Mitigating labor disruption and promoting sovereignty in a global AI market are also on the agenda.
Top executives from Alphabet, Microsoft and dozens of other businesses are slated to attend. Government leaders are expected to dine on Monday with select CEOs. And talks will include one on Tuesday by Sam Altman, chief executive of OpenAI, the maker of ChatGPT, two people involved in the summit told Reuters.
It was less clear whether the U.S. will reach consensus with other nations on AI.
Since taking office on Jan. 20, President Trump has revoked former President Joe Biden’s 2023 executive order on the technology, set in motion a repeat withdrawal from the Paris Climate Agreement and faced Congressional calls to consider new export controls on AI chips to counter rival China.
U.S. Vice President JD Vance will attend for the American delegation.
A non-binding communiqué of principles for the stewardship of AI, bearing U.S., Chinese and other signatures, has been under negotiation and would mark a big achievement if reached, said the people involved in the summit, who spoke on condition of anonymity.
They declined to detail the communiqué or elaborate if there were any points of disagreement among the would-be signatories.
The White House did not respond to a request for comment.
An official for the French presidency said the summit will give voice to countries around the world, not only the U.S. and China.
“We are showing that AI is here, that companies must adopt it, that it is a vector of competitiveness for France and for Europe,” the Élysée official said.
NO NEW AI REGULATION
Safety commitments dominated the conversation in prior global AI summits in Bletchley Park and Seoul. In Paris, creating new regulation is not on the agenda.
Reeling from red tape and a reputation for risk aversion, Europe and particularly France are eager to discuss frameworks for AI policy but not rules that could slow down their national champions, which have lagged American companies. Countries like France are evaluating how to implement the EU AI Act in as flexible a way as possible so it does not discourage innovation, the people involved in the summit said.
Instead in focus is how to distribute AI’s benefits to developing nations, via cheaper models made by the likes of France’s startup Mistral and China’s DeepSeek. The Hangzhou-based company rocked global markets last month by showing it could vie with U.S. heavyweights on human-like reasoning technology, while charging much less.
France has seized on the development as evidence that the global race to more powerful AI remains wide open.
One of the summit’s likely outcomes is that philanthropies and businesses are expected to commit an initial $500 million in capital, going up to $2.5 billion over five years, to fund public-interest projects on AI around the world, the people said.
Another is addressing the energy crunch that industry thinks is inevitable from their power-hungry AI models. A major producer of clean energy in the form of nuclear power, France wants to reconcile the world’s climate and AI ambitions.
France’s decarbonized energy and “nuclear fleet, in the context of data center installations, is an asset,” the Élysée official said. “We will most likely have announcements in this regard at the summit.”
(Reporting By Jeffrey Dastin and Elizabeth Pineau; Additional reporting by Nandita Bose and Anna Tong; Editing by David Gregorio)
Comments