Deepfake Dilemma: South Korea Eyes EU-Style AI Regulations Amid Growing Concerns

Biden’s Fake Voice Causes ‘Confusion’
US, Once Advocating Autonomy, Now Calls for Regulation Legislation
EU Enforces Strong Regulations Such as Punitive Taxes
South Korean Government: “Sanctions Possible Under Current Law”
Cautious Stance on the Need for Additional Regulations
AI Basic Law Stalled in Congress for a Year

The spread of deepfake images of pop singer Taylor Swift, synthesized using artificial intelligence (AI) technology, has led to a growing call for strong AI regulations worldwide. Amid this, the U.S. government, which has always prioritized the autonomy of AI companies, is raising the need to legislate against deepfakes. Meanwhile, the South Korean government’s stance on the need to quickly establish AI-related laws to secure AI leadership is drawing attention.

According to industry sources on the 29th, the European Union (EU) has drawn the sword of strong pre-regulation to protect traditional industries and minimize the impact of AI technology on society. At the same time, the U.S. has emphasized AI innovation based on autonomous regulation to secure the global expansion of its big tech companies and AI technology leadership.

However, in the U.S., following the dissemination of a fake voice of President Biden utilizing deepfake ahead of the November elections and the deepfake incident involving Swift, the White House and the U.S. Congress are discussing the need for AI regulation legislation.

Even in South Korea, the situation is not reassuring, which has insisted on establishing Korean-style customized AI norms referring to the U.S. and EU. Since AI is causing side effects in terms of deepfakes, fake news, and job loss, the country is also on the verge of the April general elections.

The government is closely monitoring global trends while maintaining a cautious stance on the need for additional regulations. The government argues that sanctions are possible under current law.

An official from the Ministry of Science and ICT said, “The need for AI regulation is highlighted globally due to the Taylor Swift deepfake incident, but there is no immediate move to introduce regulations.” He added, “Sanctions are possible even without AI regulation, under current laws like the Information and Communications Network Act (defamation). We are focusing on what is different and concerning when using AI, and are approaching by strengthening regulations if necessary.”

The government is also preparing a bill to mandate the display of watermarks on AI content, including deepfakes.

The Legislation on Artificial Intelligence to Make Significant Progress has been passed through the Science, Technology, Information, Broadcasting and Communications Committee of the National Assembly for nearly a year but is still pending in the full committee. The government plans to proceed with the AI law, including a legal basis to systematically support the AI industry and respond to AI bias and side effects during the 21st National Assembly session.

Likewise, South Korea has AI companies that have dominated the global market, and unlike the EU, it has companies that own AI models, so it is in a latecomer’s position. Therefore, the government settled the principle of establishing norms focusing on building trust while avoiding punitive taxes that burden companies like the EU.

Park Yoon Gyu, the second vice minister of the Ministry of Science and ICT, expressed regret that the AI-related laws have not passed the National Assembly’s threshold. He emphasized the urgency of securing AI competitiveness and hoped for the swift passing of AI-related bills.