BRUSSELS — The European Union’s biggest lenders should set out how they will tackle risks stemming from cutting-edge artificial intelligence models by the end of October. This directive comes as Europe navigates a complex landscape of digital regulation, US tech export controls, and growing concerns over digital sovereignty.
Dublin is poised to commence its six-month presidency of the Council of the EU on July 1, with tech regulation slated to be a significant component of its agenda. This presidency begins just before the French election, an event anticipated to slow down EU decision-making. Concurrently, ongoing discussions to establish a new EU-U.S. tech “dialog” are overshadowed by tensions regarding digital regulation.
Across the Atlantic, the White House’s recent restrictions on the release of new AI models are prompting cautious inquiries from tech lobbyists. This marks a notable shift, particularly given that Silicon Valley billionaires previously supported Donald Trump partly due to fears that Democrats would overregulate AI. New restrictions imposed on Anthropic, for example, have led industry advocates to express concern that the White House’s previously perceived laissez-faire approach to technology may be faltering.
Anthropic confirmed the impact of these restrictions, stating Friday evening, “The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance.” While the directive clears the path for a select group of companies and agencies to gain access to the company’s Mythos 5 model, a second advanced Anthropic model remains blocked. Washington’s export controls on Anthropic have reignited calls for Europe to accelerate the development of its own advanced AI models. Kristalina Georgieva commented on the evolving landscape, noting, “What we recognize is that Mythos is just the beginning, there will be more like it.”
Concerns over digital sovereignty are intensifying across Europe, with the continent actively working to reduce its reliance on U.S. technology for sensitive services. The U.K., for its part, has stated it has not sought a carve-out from U.S. AI export controls, opting instead for engagement with Washington.
In the face of escalating cyberattacks, infrastructure sabotage, and other hybrid threats, Ireland’s military neutrality is undergoing new scrutiny. To bolster defenses against these evolving threats, the defense alliance NATO is establishing partnerships with private companies to enhance militaries’ monitoring capabilities.
Germany is also advancing its defense infrastructure, with POLITICO having reviewed a €580 million contract that lays the groundwork for the country’s own combat cloud system. Meanwhile, Europol requires enhanced cloud infrastructure and expanded data probing powers, according to a draft mandate from the EU executive. Separately, Belgian authorities are investigating allegations of bribery involving a Chinese tech firm.
Amidst these developments, CEOs from major European companies such as Airbus, ASML, and Ericsson are engaging with European Commission President Ursula von der Leyen, advocating for deregulation and consolidation within the tech sector. Furthermore, recent report findings are likely to fuel claims by Republicans and pro-AI voices that foreign entities have attempted to meddle in the debate surrounding AI data centers.
The European Parliament president’s power play is “without precedent,” diplomats wrote in a note seen by POLITICO.
John Edwards had been subject to an independent workplace probe that found there was “a case to answer.”
