If Not 'One Rule,' Korea Must At Least Achieve 'One Stop' for AI Regulation

U.S. President Donald Trump has drawn his sword again. This time, the target is artificial intelligence regulation. The core principle is "One Rule." He signed an executive order to unify AI regulations, which had been operating separately across America's 50 states, into a single federal standard.

His statement that "there can be no innovation in a country where companies must obtain approval from 50 states every time they launch a new service" may sound like political rhetoric, but the content of the executive order is a declaration that America will not relinquish its technological hegemony. The plan to have the Department of Justice form a task force to suppress state laws conflicting with federal regulations through litigation demonstrates the resolve that "regulatory fragmentation will not be tolerated in the AI supremacy race."

The European Union is taking the opposite path. The EU's AI Act places risk-based regulation at the forefront, imposing dense obligations and prohibitions on high-risk sectors. With mandatory sandbox installations required in each member state and overlapping regulations with the General Data Protection Regulation (GDPR), corporate compliance costs have increased exponentially. Only Big Tech has the stamina to bear this burden. The result is predictable: startups are declining, venture investment is shrinking, and Europe's digital competitiveness is retreating. Standards have been strengthened, but speed has fallen.

Korea stands somewhere between the United States and the EU. Looking only at AI and autonomous driving technology, while gaps exist, Korea is evaluated as positioned right behind leading nations such as the United States and China. The problem is the regulatory framework. Korea neither has the enterprise-autonomous regulation emphasizing market discipline and post-hoc accountability like America, nor does it hold a position to export norms like Europe. If mishandled, the worst regulatory combination—EU-style strictness plus Korean-style bureaucratism—could become reality.

Complex procedures and inter-ministry conflicts are already familiar landscapes across multiple industries. The government's de facto complete ban on Initial Coin Offerings (ICOs) in 2017 is a painful example. Korean companies moved to Singapore and Switzerland to issue tokens, and innovation, jobs, and talented individuals fled overseas. Meanwhile, a distorted structure emerged where tokens issued abroad were sold to domestic investors. Korea chose regulation, but risks remained at home while opportunities were lost abroad. It was the costly price of the easy choice: "Block what we don't understand first."

AI and autonomous driving represent a far larger stage. If Korea repeats the approach of "block risks first," talent and startups will migrate to the United States and China. Domestic companies will delay investments citing regulatory uncertainty, and citizens will become dependent on AI provided by foreign platforms. Korea would become a country that imports technology while only its regulations remain domestically made.

Above all, what is needed now is a recalibration of direction. First, even if "One Rule" like America is difficult, at least "One Stop" must be achieved. Innovation cannot be discussed while leaving in place a structure where companies must wander through multiple ministries for AI and autonomous driving permits. A standing body to coordinate inter-ministry conflicts must be established, and a single window for permits must be codified.

Second, the principle of "strong regulation only for high-risk, light touch for the rest" is necessary. While strong prior regulation should apply to sensitive areas involving life and property such as healthcare and finance, it is rational to control other services through post-hoc accountability, transparency, and market competition. The moment all AI is treated as a potential "hazardous material," Korea limits its own competitiveness.

Third, regulatory sandboxes must be brought from the periphery to the center of the system. The reason Europe mandated sandbox installations is simple: without experimentation and learning, regulation for new technologies cannot mature either. Korea must also shift to a structure where all new technologies can, in principle, rapidly experiment and learn through sandboxes. Regulations should be designed based on data and evidence, not by imagining risks and blocking them.

The role of corporations is also crucial. The non-market strategy that companies must adopt in the AI era should no longer stop at responding to government regulations. This is an era where regulatory design itself determines corporate competitiveness. Companies must design and propose technology guidelines, safety standards, and data governance ahead of the government. Companies that create draft norms for the government to reference will determine the direction of regulation. Analyzing global regulatory portfolios to decide where to develop and where to launch, and establishing joint testbeds with foreign partners to secure international references—these are now tasks that corporate public affairs departments must undertake.

Trump's executive order is not simply for domestic American consumption. It is a geopolitical declaration to set the rules of AI supremacy competition the American way. To avoid falling behind in the regulatory war of the AI era, Korea must guard against excessive regulation while becoming the nation that experiments with the smartest rules first. While we hesitate, an era may arrive where others set the rules and we pay expensive usage fees. If Korea becomes a country that imports technology while only regulations remain domestically made, we can never secure our future. It is time to proceed with greater confidence, substantially overhaul regulations on cutting-edge industries including AI, and provide support so companies can compete freely.