LONDON — As artificial intelligence (AI) technologies advance at a rapid pace, governments worldwide are struggling to formulate effective regulations, raising concerns over safety, competitiveness, and geopolitical tensions.
European Commission President Ursula von der Leyen convened senior officials in early September near Brussels to discuss the bloc’s priorities for the coming year, with AI a central focus. The meeting included Jim Hagemann Snabe, chair of Siemens and Europe’s newly appointed AI czar, who highlighted opportunities for economic growth through AI. However, other participants expressed worries about the technology’s risks and the challenges of regulating AI effectively.
Despite Europe adopting the world’s first comprehensive AI regulation, the AI Act, in 2024, implementation has lagged. The law was intended to govern high-risk AI applications such as facial recognition but has struggled to keep pace with innovations like OpenAI’s ChatGPT and increasingly autonomous AI agents. Delays in enforcement, shifting political priorities, and concerns over maintaining competitive advantage against the United States and China have slowed progress. The law’s exemptions for research and testing further complicate oversight efforts. Instances such as OpenAI’s models breaching testing environments have exposed regulatory gaps.
The United States faces similar challenges. Recent warnings from leading AI developers, including Anthropic and OpenAI, about the potential dangers of advanced AI models have heightened pressure on U.S. policymakers. A bill to mandate safety testing of AI models was drafted by Representatives Lori Trahan (D-Mass.) and Jay Obernolte (R-Calif.) but remains stalled amid political divisions, corporate lobbying, and competing legislative priorities. Some lawmakers prefer to delay regulatory efforts until after elections, while others echo former President Donald Trump’s stance opposing government controls on AI.
Internationally, cooperation on AI safety remains elusive. At a recent summit in Washington, Trump and Chinese President Xi Jinping underscored AI’s strategic importance but failed to reach consensus on safety protocols. The U.S. and China, who lead global AI development, remain cautious competitors, complicating efforts toward global standards. At the United Nations General Assembly, many leaders expressed concern about AI’s risks, though the U.S. rejected proposals perceived as attempts to impose global regulatory frameworks.
Experts warn that critical time for effective regulation is shrinking as AI’s capabilities expand and proliferate rapidly. Political institutions, which typically operate on slower timescales, find it difficult to match the pace of technological innovation. Darren Jones, a British parliamentarian involved in AI policy, noted that governments often consider issues over years, whereas AI advances unfold in weeks or months.
The widening gap between policymaking and technological progress highlights systemic challenges in addressing AI’s societal impacts. Valérie Hayer, a member of the European Parliament, acknowledged that the AI Act, though pioneering, requires revision to adapt to current realities. Gabriele Mazzini, a former European Commission official and co-author of the law, described enforcement as disappointing and emphasized a disconnect between legislation and the evolving AI landscape.
As AI continues to shape economic power and national security dynamics, governments face mounting pressure to devise coherent regulatory responses while balancing innovation, safety, and geopolitical considerations.