Europe’s AI Safety Dilemma

European leaders confronted a stark choice this week: reject advanced artificial-intelligence tools and risk economic stagnation, or adopt them and become dependent on models and infrastructure owned by the United States and China. European Commission president Ursula von der Leyen announced plans to invite leading frontier labs for talks on supporting industry-led pacing of AI development.

The Guardian

The debate unfolds against the backdrop of the EU AI Act, the continent’s pioneering legislation that obliges developers of “high-risk” AI systems to report safety incidents and mitigate risks such as loss of control. While the act introduces practical rules on AI-generated content labeling, medical applications, and hiring practices, Europe still lacks a Silicon-Valley-scale AI champion comparable to OpenAI, Apple, or Google. This structural gap has limited Europe’s voice in the global AI safety conversation, a point repeatedly emphasized by policymakers and analysts.

The Guardian

She also highlighted the EU AI Act as the continent’s existing lever in the safety debate. In parallel, Frederike Kaltheuner, an adviser to the AI Now Institute, warned that the narrative of an inevitable AI frontier fuels dependency on U.S. models and chips, urging a broader, non-U.S.–China-centric approach to existential threats.

The Guardian

U.S. tech advocates contend that Europe’s stringent regulations stifle competitiveness, proposing deregulation as a remedy. Georgina Kon, a partner at Linklaters, argued that the EU AI Act is unlikely to become a global benchmark because of its heavy compliance burden, delayed implementation, and narrow territorial focus. Policy adviser Itxaso Domínguez de Olazábal of the European Digital Rights Initiative (EDRI) acknowledged Europe’s “competitiveness problem” while criticizing the narrative that blames EU law for lagging innovation.

The Guardian

If Europe continues to rely on foreign AI models, the continent risks a “sovereignty problem” where public services, education, and the economy run on infrastructure controlled by a handful of non-European firms. Conversely, investing in domestic AI capacity and expanding datacentres could reduce strategic vulnerability and give Europe a stronger footing in shaping global AI safety standards.

The Guardian

• “As for existential threats to humanity, we must first assess the actual level of threat – then act with urgency. The UN general assembly is an opportunity to address this. It is vital to move beyond the US-China rivalry narrative. There must be space for humanity between the egos of leaders like President Xi and President Trump.” — Frederike Kaltheuner, an adviser to the AI Now Institute, a research body

The Guardian

• “Europe has a competitiveness problem, that’s true,” — Itxaso Dominguez de Olazabal, a policy adviser at EDRI, a European digital rights organisation

The Guardian

Kaltheuner identified the United Nations General Assembly as a venue to assess existential AI threats and to move beyond the U.S.–China rivalry narrative. European officials have signaled intent to convene industry stakeholders and explore supranational mechanisms that could complement the EU AI Act, though concrete timelines remain unspecified.

The Guardian