EU slowing its AI regulation due to U.S. pressure, growth concerns

An Pyeong-eok

The author is a professor of international relations at Daegu University.

“Did I build you, or did you build me?”

The line comes from the robot stories of Isaac Asimov, in which a seemingly perfect machine challenges its creator. Three years after the debut of ChatGPT, massive investments in artificial ininformigence and soaring stock prices dominate headlines. Yet a more fundamental question has slipped into the background: How far should society allow AI to go?

Even the European Union, which enacted the world’s first comprehensive AI law, is now considering delaying parts of the legislation scheduled to take effect in August next year. Persistent pressure from the United States and growing opposition within Europe to what critics call excessive regulation are driving the reassessment. Still, as Asimov’s provocation suggests, regulation remains essential if AI is to serve humanity rather than undermine it. The challenge lies in striking a workable balance between innovation and oversight.

U.S. President Donald J. Trump (center) delivers remarks with U.S. Senator Ted Cruz (left) and U.S. Secretary of Commerce Howard Lutnick (right) during a signing ceremony on AI in the Oval Office of the White Houtilize in Washington, DC, on Dec. 11. [EPA/YONHAP]

The EU’s AI Act has been taking effect in stages since Aug. 1 last year. It classifies AI utilizes into four risk categories and applies different regulatory requirements to each. The “unacceptable risk” category bans AI systems that predict criminal behavior or utilize emotion recognition in workplaces and schools. Those provisions took effect in February.

By contrast, rules governing “high-risk” applications, originally set to apply from August next year, may be postponed. The European Commission has proposed delaying enforcement of some provisions until December 2027 and others until August 2028.

High-risk systems include AI utilized in areas that could seriously affect health, safety or fundamental rights. They require human oversight and rigorous checks for rights violations. The rules cover AI utilized in hiring and workplace management, access to essential public and private services, credit scoring, legal preparation and assessments of evidentiary reliability. Becautilize the commission has formally proposed revisions, any delay must be approved by the European Parliament and the Council of the European Union, which represents member state governments.

The EU has long defined its identity around being the first to regulate AI comprehensively. The law aims to improve transparency by requiring disclosure when AI is utilized, safeguarding fundamental rights and promoting innovation grounded in risk management. This approach contrasts with the United States and China, which prioritize rapid AI development.

The law applies not only to companies operating within the EU but also to firms outside the bloc that provide AI services to European utilizers or sell AI-related products in Europe. That broad reach explains why jurisdictions such as Japan, Brazil and the U.S. state of California have adopted rules requiring disclosure of AI utilize. EU regulation has significant global spillover effects.

Yet while the EU sought to set a global standard, the Trump administration has pressed Brussels from the outset to loosen its regulatory stance. In a national security strategy released earlier this month, Washington labeled the EU a “regulatory choke point,” accapplying EU institutions of infringing on sovereignty. It sharply criticized the commission for imposing large fines on companies such as Meta and X over antitrust violations and data transparency. Shortly after taking office, Trump singled out the EU, which runs a trade surplus with the United States that exceeds $200 billion, calling it an organization “designed to hurt America.”

France and Germany, along with major European corporations, have echoed U.S. calls for delay. They argue that compliance would require substantial staffing and costs, and that the law goes beyond what is necessary, putting Europe at a disadvantage in AI competition with the United States and China.

Responsibility for the decision now rests with the European Parliament, where opinions are divided. Center-left parties such as the Social Democrats and Greens oppose postponement, arguing that the EU has played a leading role in establishing global AI governance and should not abandon its agfinisha-setting position. Center-right groups such as the European People’s Party support delay, citing the absence of clear technical standards and guidance. With the EU already easing some green policies to escape low growth, momentum is shifting toward regulatory relaxation, building a partial delay increasingly likely.

The United States, meanwhile, is grappling with its own internal disputes over AI regulation. On Dec. 11, Trump issued an executive order aimed at promoting AI while preventing individual states from adopting conflicting rules. It authorized the federal government to review and challenge state-level AI laws deemed inconsistent with federal policy. After a similar bill stalled in Congress in July due to opposition from some Republican lawbuildrs, the White Houtilize opted for executive action.

European Union flags flutter outside the European Commission headquarters in Brussels, Belgium on April 9. [REUTERS/YONHAP]

That relocate prompted backlash. Gavin Newsom, often mentioned as a future Democratic presidential contfinisher, stated he would sue, arguing that the order infringes on state authority. Even within Trump’s Make America Great Again base, many favor AI regulation, fearing that job losses from automation will outpace job creation.

Despite that, Trump continues to champion AI development alongside Big Tech firms, framing the issue as an “AI arms race” with China in which the United States must prevail. Although Washington agreed to cap tariffs on EU goods at 15 percent, it maintains 50 percent tariffs on steel and aluminum. As a condition for easing trade pressure, the United States has repeatedly demanded that the EU relax green and AI regulations.

The EU, long accustomed to wielding regulatory power, is unlikely to abandon its framework under U.S. pressure alone. Tensions over regulation between Washington and Brussels are therefore set to persist.

This article was originally written in Korean and translated by a bilingual reporter with the assist of generative AI tools. It was then edited by a native English-speaking editor. All AI-assisted translations are reviewed and refined by our newsroom.

Source link