TL;DR
Brussels is advancing InvestAI and cookie-consent reforms at the same time, highlighting a gap between Europe’s digital rulemaking and its AI production base. Commission figures cited in the source material put non-EU digital imports at about €264 billion a year, while Thorsten Meyer AI argues the bloc’s AI plan is still small next to US and Chinese capacity.
The European Commission has moved to fund AI gigafactories through its InvestAI plan and reduce cookie-banner friction through digital simplification proposals, a dual push that puts Europe’s digital policy gap in focus: the bloc is trying to build AI capacity while still relying heavily on non-European cloud, models and capital.
According to Commission figures cited in the source material, InvestAI is meant to mobilize €200 billion, including €50 billion in public money and an expected €150 billion from private sources. About €20 billion is ring-fenced for AI gigafactories, with EU funds covering up to 17%, and operational compute is expected in 2027-28.
The same analysis contrasts that plan with the scale of spending by US hyperscalers. It cites FT-compiled estimates that Amazon, Microsoft, Google and Meta are together set to spend about $700 billion in capital expenditure in 2026 alone, while the Stargate project is described as a $500 billion effort.
On the interface side, the Commission’s digital simplification package seeks to reduce cookie-banner friction through one-click choices and browser-level preferences. The Commission says the changes could save businesses €800 million a year. Legiscope, a consent-management vendor, estimates EU users spend about 575 million hours a year dismissing cookie banners, but that figure is a vendor estimate rather than a confirmed official measure.
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 Leverage Problem
The policy question is not whether privacy or AI safety matter. The issue is whether Europe can shape the next technology stack if it does not own much of the infrastructure behind it. The source material cites Commission data showing about €264 billion a year spent importing non-EU digital products, more than 80% reliance on non-EU digital systems, and about 70% of EU cloud held by AWS, Google and Microsoft.
That dependency affects companies, public services and researchers that need compute, cloud tools and advanced models. Rules can change how technology reaches users, but they do not by themselves create frontier labs, cheap power contracts, scale-up finance or model distribution. Thorsten Meyer AI frames the problem as a gap between regulating the visible interface and building the industrial engine underneath.
AI gigafactory construction kit
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
From Consent Rules to Compute Plans
The cookie-banner system is tied mainly to Article 5(3) of the older ePrivacy Directive, which governs storing or accessing information on a user’s device, rather than to the GDPR alone. Studies of real-world banners have found widespread compliance problems, including dark-pattern designs and unclear purpose descriptions. One analysis of roughly 400 banners found about 89% breached rules in some way, according to the source material.
Europe has also moved early on AI governance through the AI Act, while the leading commercial and open model ecosystems have largely formed in the United States and China. The source material identifies Mistral as Europe’s main serious LLM contender, but says its strongest open model remains behind leading US systems on hard reasoning benchmarks and that its war chest is far smaller than single large US fundraising rounds.
The structural limits cited in the analysis include a thinner scale-up capital market, pension funds that play a smaller role in venture financing, industrial electricity costs around double US levels according to ACER, slow grid buildout, and talent moving toward markets with more compute, compensation and capital.
“InvestAI is intended to mobilize €200 billion for AI investment, including a €20 billion fund for AI gigafactories.”
— European Commission
cookie consent management software
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Open Questions on EU Capacity
It is not yet clear whether InvestAI will draw the full private funding the Commission hopes for, where the gigafactories will be built, or whether they will have the affordable power needed to compete with US and Chinese infrastructure. It is also unclear whether the cookie-banner reforms will improve user control or mainly reduce visible friction.
AI rankings can change quickly as new models ship, so Europe’s position may shift if Mistral or other labs release stronger systems. The source material’s claims about benchmark placement and model usage should be read as a snapshot from late June 2026, not a settled long-term ranking.
browser extension for cookie preferences
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Gigafactory Decisions Move Forward
The next tests are practical: formal progress on the digital simplification package, funding commitments for InvestAI, site selection and procurement for gigafactories, power agreements, and the first operational compute in 2027-28. Policymakers and industry will be watching whether the plan produces competitive European models, or whether Europe remains mainly a regulator and buyer of systems built elsewhere.
AI development hardware
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
What is the actual news development?
Brussels is moving ahead with AI investment plans through InvestAI while also trying to simplify cookie-consent rules. The pairing has renewed debate over whether Europe has focused more on regulating digital interfaces than building core AI infrastructure.
Does this mean EU privacy rules failed?
No. The article does not say privacy protections are wrong. It says the cookie-banner system has produced heavy friction, uneven compliance and limited user control, while the underlying AI infrastructure race has moved faster outside Europe.
How far behind is Europe in AI?
The source material says Europe has one main LLM contender, Mistral, while the leading closed frontier systems are in the United States and several strong open-weight systems are coming from China. Exact rankings remain fluid because model releases and benchmarks change quickly.
What would change the picture for Europe?
The analysis points to affordable sovereign compute, stronger scale-up capital, competitive power costs, open-weight model distribution and better talent retention. Funding alone may not close the gap if those constraints remain.
Source: Thorsten Meyer AI
