TL;DR
Thorsten Meyer AI reports that Europe is preparing a major AI funding push while remaining heavily dependent on non-EU cloud, compute and frontier models. The analysis says the EU’s regulatory strength has not translated into ownership of the infrastructure and capital needed to compete in advanced AI.
Europe is moving to spend more public money on artificial intelligence infrastructure, but a new Thorsten Meyer AI analysis says the bloc remains far behind the United States and China in frontier models, cloud capacity, venture capital and affordable power, making its bid for AI sovereignty uncertain.
The analysis points to the European Commission’s InvestAI plan, described as an effort to mobilize €200 billion, including €50 billion in public funding and €150 billion in expected private investment. It says only about €20 billion is ring-fenced for AI gigafactories, with EU funds capped at 17%, and that meaningful compute capacity is expected in 2027 or 2028.
By comparison, the report cites financial data suggesting that the four largest U.S. hyperscalers are on track to spend about $700 billion in capital expenditure in 2026 alone. It also cites the $500 billion Stargate project as a sign of the scale gap Europe faces.
Thorsten Meyer AI says Europe’s strongest AI lab, Mistral, remains an important but undercapitalized contender. The analysis describes Europe as having no entrant in the most restricted frontier-model tier and says the bloc’s strongest advantages are price and jurisdiction rather than raw capability.
Europe regulated the interface and forgot the engine
The cookie banner is the most-used European software of the decade. While Brussels perfected the consent pop-up, the frontier was built elsewhere — and now, in H2 2026, Europe wants to buy back in without changing what put it on the outside.
This isn’t about whether privacy or safety matter — they do. It’s that Europe mistook regulating the interface for having a seat at the table. You can’t grant your way out of a structural problem while keeping the structure — the laws, the capital gaps, the energy costs, the talent drain all left untouched. The fix isn’t another framework: it’s open weights as a product, sovereign compute on affordable power, real capital plumbing — and to stop mistaking a check for a strategy.
Europe’s AI Sovereignty Gap
The issue matters because AI systems are becoming part of cloud services, defense planning, software development, industrial automation and public administration. If European companies and governments depend on non-EU infrastructure and models, policy choices made in Brussels may have limited effect on the systems Europe actually uses.
The analysis says Europe spends about €264 billion a year importing non-EU digital products, relies on non-EU digital infrastructure for more than 80% of its digital stack, and has around 70% of its cloud market served by AWS, Google and Microsoft. Those figures are attributed in the source material to the European Commission and related policy reporting.
The report’s central claim is not that privacy or AI safety rules are unnecessary. It argues that rulemaking alone has not produced the capital markets, energy capacity, chips, cloud infrastructure or talent base needed to build frontier AI at scale.
AI infrastructure server racks
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
The analysis uses Europe’s cookie-consent system as a symbol of the bloc’s digital policy record. It cites Legiscope, a consent-management vendor, estimating that EU internet users collectively spend about 575 million hours a year dismissing cookie banners. The report treats that as a rough vendor estimate, not a confirmed public statistic.
It also cites research finding that many cookie banners do not comply with consent rules, including one analysis of roughly 400 banners that found about 89% had legal problems, such as dark patterns or unclear purposes. The legal trigger for many banners comes from the ePrivacy Directive’s Article 5(3), which covers storing information on a user’s device.
The European Commission’s Digital Omnibus proposal is cited as an attempt to reduce cookie-banner friction through one-click choices and browser-level preferences. According to the source material, the Commission has said those changes could save businesses about €800 million a year.
European AI development hardware
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Funding Impact Still Unclear
It is not yet clear how much private capital InvestAI will attract, how quickly gigafactory capacity can come online, or whether European power prices and grid delays will limit AI infrastructure buildout. The report also does not establish whether Mistral or another European lab could narrow the capability gap through new model releases.
Some of the performance comparisons in the source material rely on benchmark trackers and usage rankings, which can change quickly and may not reflect enterprise adoption, safety performance or domain-specific value. The analysis presents them as a snapshot as of late June 2026.
AI compute capacity hardware
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Brussels Faces Execution Test
The next test is whether the European Commission and member states can turn AI funding plans into operational compute, cheaper power access, larger growth-capital pools and competitive model development. The report says Europe’s challenge is no longer only to write rules for digital markets, but to build and finance the systems those rules are meant to govern.
Readers should watch for final InvestAI commitments, gigafactory site decisions, private co-investment levels, energy-policy changes and new European model releases through 2027 and 2028.
AI gigafactory equipment
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
What is the main development in this story?
A Thorsten Meyer AI analysis says Europe’s new AI investment push exposes a deeper problem: the bloc regulates digital technology heavily but still depends on non-EU companies for much of the cloud, compute and frontier AI stack.
Is the EU spending €200 billion directly on AI?
No. The source describes InvestAI as an effort to mobilize €200 billion, including €50 billion in public funding and €150 billion in hoped-for private investment.
Why are cookie banners part of an AI story?
The analysis uses cookie banners as an example of Europe’s focus on user-facing regulatory interfaces. It argues that this did not translate into control over the deeper infrastructure now driving AI power.
What remains uncertain?
It is still unclear how much private investment will arrive, when new compute will be operational, whether power costs can be reduced, and whether European AI labs can close the capability gap.
Source: Thorsten Meyer AI
