BURSTHow much money might it take for Europe to close its AI gap and catch up with world leaders such as the U.S. and China? If you ask top European Union (EU) officials, they might say €200 billion.
That’s the amount that EU Commission President Ursula von der Leyen referenced when she launched an overarching plan called InvestAI earlier this year intended to galvanize technology companies, academia, commercial users, and financiers into a cooperative push to advance the state of the art and the uptake of AI.
Speaking at the Artificial Intelligence Action Summit in Paris, von der Leyen articulated the benefits the technology could bring: “AI will improve our healthcare, spur our research and innovation, and boost our competitiveness,” she said. “We want AI to be a force for good and for growth.”
The €200-billion InvestAI initiative “will enable all our scientists and companies—not just the biggest—to develop the most advanced very large models needed to make Europe an AI continent,” von der Leyen said. Among other intentions, her plan calls for building four “AI gigafactories,” enormous computing centers that would have the power to develop ambitiously large AI models required for major breakthroughs in areas such as science and medicine.
Yet in a tacit acknowledgment that Europe will have to play catch-up, she noted in announcing the plan, “Our approach still needs to be supercharged.”
That is indeed the case, given the yawning gap between the state of AI in Europe versus the U.S. and China.
European Standouts
Europe has its AI stars. Two-year-old Paris startup Mistral AI, for example, gained early and rapid success by creating an AI model with fewer computing resources than its U.S. rivals. Other notable startups have included France’s AI model company Poolside; the U.K.’s Wayve, specializing in AI for driverless vehicles; and Germany’s DeepL, which makes AI translation software, and Black Forest Labs, which specializes in image generation.
Another German AI startup, Munich-based Helsing SE, has staked out a strong presence in developing AI for defense applications, including autonomous controls of military craft. Its Centaur AI agent recently took over the piloting of a Saab AB fighter in two different test flights over the Baltic Sea. Helsing has expanded into making its own AI-driven drones and submarines. In February, Helsing and Mistral struck up a partnership in which Mistral AI models will help Helsing powered weapons make decisions such as what to strike and when.
And in 2024 Google’s London-based DeepMind unit developed an AI model called AlphaFold2 that won its creators Demis Hassabis and John Jumper a share of the Nobel Prize in Chemistry 2024 for predicting proteins’ complex structures, a development heralded for its usefulness in medical and drug development circles.
That suggests a wealth of intelligence knocking around the EU and wider Europe (the U.K. exited the EU in early 2020) in the field of artificial intelligence.
A Long Way to Go
However, compared to the U.S. and China, there is a major chasm in both the development and deployment of AI.
Stanford University’s Institute for Human-Centered Artificial Intelligence (HAI) keeps a close eye on these things. Its 2025 AI Index Report sized up all sorts of AI developments in 2024, including the production of AI models—essential to the deployment of AI in that they are the programs, such as OpenAI’s ChatGPT and Google’s Gemini, that are trained to perform specific tasks without humans.
From a European perspective, the rankings were dismal.
“The United States continues to be the leading source of notable AI models,” the HAI report said. “In 2024, U.S.-based institutions produced 40 notable AI models, significantly surpassing China’s 15 and Europe’s combined total of three.”
One reason for the gaping difference: money.
“U.S. private AI investment hit $109.1 billion in 2024, nearly 12 times higher than China’s $9.3 billion and 24 times the U.K.’s $4.5 billion,” the HAI report noted. “The gap is even more pronounced in generative AI, where U.S. investment exceeded the combined total of China and the European Union plus the U.K. by $25.4 billion, expanding on its $21.8-billion gap in 2023.”
In other words, the striking disparity in investment has been growing more pronounced. To break that down further, HAI noted that while private investment in European AI has grown 60% since 2023, the U.S., which was already way ahead, “has seen a significant increase (+50.7%) during the same period—and a +78.3% increase since 2022.”
Can Anyone Spare €200 Billion?
That’s where von der Leyen’s €200 billion could come to the rescue.
From a financial perspective though, the big unanswered question is where all those billions will come from. The EU is not funding it per se; rather, as von der Leyen made clear, it wants to “mobilize” the €200 billion through “public-private partnership.” The European Investment Bank, which describes itself as the lending arm of the European Union, will play a role, as will member-states.
Yet much of the funding will have to come from private sources. Multiple requests from Communications to speak with EU officials about progress in lining up contributors, be they public or private, were declined. In an email, a spokesperson referred to the important role that a group called the EU AI Champions Initiative will play, noting that it had “pledged a €150-billion investment from providers, investors, and industry.”
The EU AI Champions also declined Communications‘ request for an interview. A Champions spokeswoman said via email that “the key capital allocators stand ready to activate €150 billion.” She gave no further details.
We Are the Champions
The EU AI Champions Initiative is a group of about 70 financial, technology, and industrial companies that is not an altogether European entity. It is spearheaded by Cambridge, MA-based General Catalyst, a venture capital and private equity firm, with New York, NY-based investment outfits Blackstone Inc. and KKR, and others also playing key roles.
The initiative has landed some big European names from the industrial and business worlds, such as Airbus, Mercedes Benz, Volkswagen, Porsche, Saab, Philips, Siemens, L’Oréal, Renault, Lufthansa, SAP, Dassault, BNP Paribas, and Deutsche Bank, plus various AI technology outfits.
Whether more than 70 (the number is growing) companies, some of them direct competitors, can get together and cooperate on developing strategic technology remains to be seen. Business history is littered with failed consortia. Though it’s still early days for the Champions.
However, there is some skepticism towards the broad €200-billion InvestAI program.
“I think it’s highly bureaucratic,” said Alexandra Mousavizadeh, CEO and co-founder of Evident, a London AI consulting and research firm. “It’s a classic European, ‘We’ve got to have some sort of strategy, and then we’ll think about it, we may spend some money on it.’ I am very skeptical. I don’t think that they understand the urgency. I don’t think they’re deploying it fast enough. I don’t see any evidence of that.”
Adoption of AI by European companies across industrial sectors lags way behind that of the U.S., Mousavizadeh said. Stanford HAI was more upbeat in its adoption assessment.
It’s All in the Mindset
That’s not to say that companies won’t rise to the occasion, with or without the EU plan. But for that to happen, European companies must change their mindsets regarding how to use AI, said Mousavizadeh.
Evident specializes in tracking AI in the financial industry where, she noted, U.S. companies such as JP Morgan Chase, Capitol One, Wells Fargo, and Goldman Sachs started focusing on AI in 2016, driven by a notion of business transformation. JP Morgan Chase CEO Jamie Dimon famously referenced the importance of AI in a 2017 letter to shareholders, a development that Mousavizadeh interprets as turning the outfit into an AI company.
“The European organizations are only starting to think that way now,” she said. “They are significantly behind. Adoption and ROI are much higher in U.S. banks.”
It’s not just banks. The adoption curve in Europe is generally behind across most sectors. Mousavizadeh faults the lack of an ecosystem, with Europe having far fewer AI development companies and specialists in how to integrate AI into a business.
“All of the know-how and providers of tech solutions, all of the solutions are U.S. and all of the LLM (large language model) providers are in the U.S., bar Mistral in France,” she noted.
“It means the ecosystems are weak; the proximity of business to the development of AI means the deployment is slower. You have data scientists and data engineers in Europe, which is fine, but you don’t have serious AI product management, or AI software implementation talent, which is needed to do GenAI or agentic AI.”
There are bright spots in the European ecosystem, Mousavizadeh noted. With Mistral, Paris has built up a good cluster of organizational users, for example.
Then again, Mistral, which shot to fame by doing more with less, was outdone at the same game by China’s DeepSake last year.
Europe has a long way to go to catch up, and to maintain its advances. Whether €200 billion can do the trick—and whether €200 billion even exists for that purpose—is a question that even the most advanced AI model cannot yet answer.
But with or without InvestAI, Europe has AI strong suits.
Facing pressure to build up its defense capabilities, for example, the military sector could become a hotbed of AI pursuits such as at Helsing, which has provided AI-equipped drones for deployment in Ukraine.
Mark Halper is a freelance journalist based near Bristol, England. He covers everything from media moguls to subatomic particles.