Europe moves towards tech sovereignty to 'protect citizens and defend interests'
It is part of a drive for Europe to be known as an AI continent and a global leader in this technology
New plans have been unveiled to shift Europe towards technological autonomy in a move to protect citizens and defend interests.
Listen to this article
The European Commission has introduced measures to lessen the European Union's reliance on tech firms based outside the bloc, particularly as demand for computing capacity rises with the spread of artificial intelligence (AI).
As part of the European Technological Sovereignty Package, the Commission is introducing two pieces of legislation: the Chips Act 2.0 and the Cloud and AI Development Act - as well as the Open Source Strategy and a Strategic Roadmap for Digitalisation and AI in Energy.
This will develop Europe's strengths in mainstream chips and building capacity in cutting-edge semiconductor technologies, whilst tripling data centre capacity across the continent.
Read more: Starmer ‘100% behind’ MP suing Elon Musk’s xAI over Grok images
Ursula von der Leyen, Commission President, said: "We cannot afford to depend on others for the technologies that keep our hospitals running, our energy grids stable and our services secure.
"This is about protecting our citizens, defending our interests and making our own choices. Europe has the talent, the research excellence, the industrial base and the Single Market. Together, we must turn these strengths into technological sovereignty."
It is part of a drive for Europe to be known as an AI continent and a global leader in this technology.
The Commission is expected to launch a call for 'AI gigafactories' in July as well as a consultation to finance the push for development and research in the space.
James Milnes, founder of British sovereign-AI firm MissionOpsAI and a former Senior Military Officer, said: "What Brussels set out this week is less a ban than a turning point.
"For the first time, European law would treat sovereignty as a condition of public procurement rather than a preference; ranking technology providers by how much of their operation sits under European control, and obliging buyers in banking, energy, healthcare and defence to weigh that alongside price. It is a proposal, not yet law, but the direction is unmistakable."
The Commission will introduce a framework that assess cloud and AI sovereignty to protect critical applications and sensitive data.
It says that governments relying on foreign infrastructure for military communications, critical national infrastructure, and public administration could see their sovereignty undermined.
Further, sovereign tech allows for better regulation as lawmakers are able to audit, inspect, and intervene in these systems, whereas this isn't open to them if the technology they are using operates under foreign laws.
Mr Milnes added: "The uncomfortable truth underneath it was admitted under oath a year ago, when Microsoft’s own counsel told the French Senate it could not guarantee that European data would be shielded from US authorities.
"A company subject to US law must comply with US law, wherever its servers happen to sit. No amount of ‘European region’ branding changes that, and once it is understood, the policy follows logically."
Earlier this year, the UK announced a £500 million sovereign AI unit to support homegrown companies - the first beneficiaries include Callosum, Isomorphic Labs, and Ineffable.
The new unit will both invest in British AI firms and provide them with access to supercomputers, research support and help bidding for government contracts.
In March, the Chancellor told an audience in the City of London that the UK would pursue “the fastest adoption of AI in the G7”, arguing a thriving AI industry was “a critical part of our national resilience”.
"This is not only a continental story," Mr Milnes said.
"Britain is making the same bet from the other direction, through the Government’s £500 million Sovereign AI effort and the ambition to be, in the Prime Minister’s words, an AI maker rather than an AI taker.
"The question facing Whitehall and Brussels alike is the same: how do you build and buy artificial intelligence you can actually trust with the most sensitive functions of the state?
"My concern is that the debate too often stops at where the data sits. Location matters, but it is not the whole of sovereignty.
"What matters just as much is whether a system can be governed, held to account and independently verified: whether you can prove, after the event, exactly what it did and why. Sovereignty you cannot audit is a slogan, not a safeguard. The countries and companies that take that seriously will set the standard everyone else has to meet."