France’s newsy week hasn’t been limited to a surge for the far right in the European Parliamentary elections, French President Emmanuel Macron’s snap dissolution of Parliament and a très-dramatic French Open final. Mistral AI, the most significant non-American challenger in the race to develop powerful artificial intelligence systems, closed its Series B funding round to the tune of $640 million. The company, co-founded by alumni of Google DeepMind and Meta, now has a total valuation of $6 billion, raising money from Andreessen Horowitz, Nvidia, IBM, Samsung and a slew of European investors. It’s another huge win for a company that seems as well-poised as any to seriously challenge Silicon Valley stalwarts like OpenAI and Anthropic. But tech works differently in Europe than it does here — not least because politicians and regulators have real sway over the industry. For a company like Mistral, in a nation as proud as France, it simply matters more at the geopolitical level: Co-founder Cedric O mused (via Google Translate) in his most recent Medium post that American tech “domination has massive consequences in technological, economic and geopolitical terms, which must be understood in their entirety in order to judge Europe's backwardness, the risks weighing on our continent, its potential, too, and the the imperative obligation placed upon it to rise to this challenge, failing which its economic and political sovereignty will inevitably be threatened.” In other words: Europe can compete with Silicon Valley, but only if its government(s) back its private sector all the way. Mistral is a national, if not continental champion, and has benefited massively from that backing thus far. POLITICO’s Mohar Chatterjee and Gian Volpicelli reported last year that its odd position as the hopeful titan in a land that usually puts the clamps on tech giants has led it to explicitly seek the EU’s institutional support. So far, that’s working: Mistral lobbied hard for and ultimately received industry-friendly 11th-hour concessions to the text of the AI Act. O is politically connected himself, but his ties are to the party now in danger of losing power in France. He’s a member of Macron’s centrist Renaissance party and its former Secretary of State for the Digital Sector, meaning he (and presumably Mistral) have an obvious dog in July’s French election fight if they want to ensure the friendliest possible audience in Paris. If Macron remains president until 2027 but is saddled with a National Rally-led parliament, the ensuing political chaos could make choppy the heretofore smooth political waters the company has faced. There’s also an argument that the AI debate in Europe lies elsewhere: The EU’s AI Act is already on the books, meaning that the laws most relevant to an AI contender like Mistral will now be enforced by EU bureaucrats beyond the reach of the political chaos that could roil the continent (and France itself). So to figure out what a more right-leaning Europe would mean for companies like Mistral and the overall tech world, it’s worth looking at how they’ve treated it so far. For one, the right just tends to use the technology more freely than its counterparts: POLITICO’s Gian Volpicelli pointed out in today’s EU Influence newsletter (for Pro subscribers) that the most high-profile uses of generative AI in this year’s elections came from the right. There were Dutch right-wing leader Geert Wilders’ eerily propagandistic images of happy Dutch “families” posted to X. Critics accused Matteo Salvini, leader of Italy’s far-right League, of using AI to generate an image of a pregnant trans man to stoke fear among social conservatives. A comprehensive review of the data by the Atlantic Council’s DFRLab showed that the European Parliament’s far-right ID coalition was far more likely than than their opponents to use generative AI. All of this tracks with data in the United States, where even those on the right who fear AI are shown to be less willing to regulate its use. And then for another datapoint, there’s the actual roll call for the outgoing European Parliament’s vote on the AI Act: All but one of the members of ID voted for the bill (along with an overwhelming 523-vote majority), with the majority of dissent coming from left-leaning groups who said the bill didn’t go far enough in restraining companies … like Mistral. Given the European right’s clear affinity for generative AI overall and desire to protect a hazily-defined European “sovereignty,” it’s not difficult to imagine the France-first company navigating the winds of a more right-leaning Europe more smoothly than other continental giants. And now that the AI Act is in the books, in the near future the European Parliament seems more likely to turn its attention to less flashy, easier-to-agree-on policy goals like protecting children online, as the outgoing Dutch Digital Minister Alexandra van Huffelen told POLITICO’s Morning Tech Europe today. Meanwhile Mistral continues to forge ahead amid its happy economic news, deploying new customization features and ramping up its civic participation by joining the EU Internet Forum. If Cedric O’s political party finds itself on the wrong end of France’s high-pressure upcoming Parliamentary election, at least he can likely console himself with some positive business news.
| 
Comments
Post a Comment