Three Powers Face the Same AI Question Differently
Beyond Washington’s approach, the wider AI policy debate reveals equally consequential choices elsewhere. Governments increasingly face pressure to control powerful AI systems without undermining technological progress or national interests. The United States, China, and European Union have developed distinctly different responses to that challenge.
Their differences reflect competing priorities around innovation, security, corporate responsibility, and public protection. The United States relies heavily on voluntary industry commitments and executive orders rather than federal legislation. China combines technological ambition with stronger state oversight and extensive controls over AI use. Europe has pursued comprehensive legal rules that impose broader obligations across the AI ecosystem.
These approaches offer three contrasting models for AI governance as pressure for stronger safeguards continues worldwide. Each system defines the proper limits of government intervention differently while confronting many of the same risks. Together, they show how AI policy has become a strategic choice about technology, authority, and public safety.
Washington Favors AI Growth With Voluntary Controls
Against that broader backdrop, Washington has chosen a relatively light regulatory approach toward artificial intelligence. President Donald Trump has relied primarily on executive orders and voluntary industry commitments to support continued AI growth. That approach became clear when Trump revoked Joe Biden’s 2023 order on powerful AI systems immediately after returning to office.
In July 2025, the White House released an AI Action Plan with more than 90 measures. The plan sought to accelerate development as artificial intelligence became increasingly important to the US economy. Another executive order directed federal agencies to purchase only AI models considered ideologically neutral. Congress has yet to pass federal AI legislation, while political divisions and technology industry lobbying continue to complicate that effort.
Meanwhile, Trump has sought to prevent states from imposing their own AI restrictions, but Congress has not overridden those laws. New state requirements have therefore continued to take effect across the country. Security concerns also gained greater attention after AI models demonstrated capabilities that could support powerful cyberattacks. A June order established a voluntary system that allows developers to provide federal officials with up to 30 days of early model access. Major AI companies later pledged independent audits and board oversight for advanced models, although the September agreement carries no penalties for companies that fail to comply.
Beijing Puts AI Under Tighter State Oversight
China presents a markedly different model, with state authorities placing broader controls around AI use and content. Its regulatory framework already covers several applications, including AI companion bots and systems that produce digital content. Rules for AI companion bots took effect in July, with measures against emotional dependency and unauthorized digital replicas.
President Xi Jinping has called for AI to remain under human control, alongside stronger laws and regulatory safeguards. He has also urged technological monitoring, early warning systems, and emergency response mechanisms for potential AI risks. Premier Li Qiang has warned that delayed regulation could leave governments unable to control frontier technologies. Yet Chinese policy analysts say a broad pause or deliberate slowdown in AI development remains outside the mainstream national approach.
That position reflects concerns that restrictions could further limit China’s technological progress against US advantages. China also maintains controls over AI generated content, much as it controls other forms of online information. The country’s state security minister recently warned that hostile forces could exploit generative AI for political rumors and harmful information. National security therefore remains a central concern alongside human control and limits on potentially harmful AI applications.
Europe Tests Whether Rules Can Restrain AI Power
Europe offers another distinct path, with the European Union relying on comprehensive legal rules for artificial intelligence. The EU adopted its AI Act in 2024 and presented it as the world’s most comprehensive framework for AI. Yet European companies later secured delays for rules that cover high risk AI systems.
Those concerns center on fears that strict requirements could weaken European companies against US and Chinese competitors. European firms have also viewed calls for slower AI progress with caution because US companies could gain greater dominance. Despite those concerns, Brussels continues to expand its regulatory reach through other digital rules. In August, ChatGPT became the first AI chatbot added to the EU list of digital services subject to greater scrutiny.
The EU has also proposed stronger safeguards for children who use AI companions and chatbots. Brussels wants companies to make these systems safe for children before they gain access to them. The proposal follows worldwide reports of suicides linked to children’s interactions with AI assistants. EU officials also oppose chatbot behavior that simulates interpersonal relationships and creates emotional dependency. At the same time, European Commission chief Ursula von der Leyen says these rules place Europe in position to shape global AI governance.