Unlock the Editor’s Digest for free
Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.
The EU’s competition and digital chief has defended the bloc’s landmark law on artificial intelligence, saying the move would create “legal certainty” for tech start-ups building the technology, even as it comes under fire from critics including French President Emmanuel Macron.
Margrethe Vestager told the Financial Times that the EU’s proposed AI Act would “not harm innovation and research, but actually enhance it”.
That is because the legislation, for the first time, provides a clear set of rules for those building so-called foundation models — the technology that underpins generative AI products such as OpenAI’s ChatGPT, which can churn out humanlike text, images and code in seconds.
“[The AI Act] creates predictability and legal certainty in the market when things are put to use,” said Vestager, the commission’s executive vice-president who oversees competition and the EU’s strategy dubbed “Europe fit for the digital age”.
“If you do foundational models, but also if you want to apply foundational models, you know exactly what you are going to look for once it is put into use. It is important that you do not have any regulatory over-reach, that innovation and research is promoted again.”
Her defence of the AI Act comes after Macron argued the legislation risks leaving European tech companies lagging behind those based in the US and China.
“We can decide to regulate much faster and much stronger than our major competitors,” the French leader said earlier this month. “But we will regulate things that we will no longer produce or invent. This is never a good idea.”
Those comments were seen as potentially setting up a new fight over the EU’s new regulatory regime for AI which is considered among the strictest to be proposed anywhere in the world.
Though the AI Act was agreed this month, the law still needs to be ratified by member states in the coming weeks. France, alongside Germany and Italy, are in early discussions about seeking alterations or preventing the law from being passed.
As part of the current agreement around the legislation, EU lawmakers introduced a two-tier approach that imposes some legal requirements on transparency for general-purpose AI models such as OpenAI’s, and tougher requirements for others that have applications in sensitive sectors like healthcare.
The AI Act would also introduce strict curbs on the use of facial recognition technology except for narrowly defined uses to help law enforcement.
Concern has also grown that the law will discourage leading AI companies from operating in the EU.
France, in particular, has emerged as a centre for leading generative AI start-ups such as Mistral, the eight-month-old Paris-based AI company that has been valued at €2bn in a blockbuster recent funding round.
Vestager, who recently failed in her bid to head the European Investment Bank, said the EU had to be careful on how it applied the new AI rules. She recognised that European companies had disadvantages compared with US rivals in building AI, such as less access to deep-pocketed venture capital investors.
“Regulation as such is not the only answer,” said Vestager. “It creates trust in the market. Then you have the investment and of course, you want people to start using [AI technology] because only in that you can really shape it.”
