South Korea’s Pioneering Regulations on Artificial Intelligence
South Korea has become the first country in the world to implement laws regulating the operation of artificial intelligence (AI) systems. As of now, images created with AI must be clearly labeled, and individuals in critical sectors will be required to oversee the functioning of these systems, as reported by the BBC.
The South Korean authorities assert that these new laws aim to strengthen public trust in AI technologies and enhance safety measures. The Ministry of Science and ICT has characterized this legislative framework as the second of its kind globally.
Global Context of AI Regulation
Disagreements persist around the globe regarding the regulation of AI. For instance, the USA favors a more lenient approach to encourage innovation, while China has introduced a series of rules and proposed the establishment of a body to coordinate international regulations.
Key Provisions of the New Legislation
The new South Korean laws stipulate that companies must ensure human oversight over “high-performance” AI technologies, particularly in crucial areas such as:
- Nuclear safety
- Drinking water production
- Transportation
- Healthcare
- Financial services, including credit assessments and loan approvals
Additionally, companies are mandated to notify users beforehand about products or services that utilize high-performance or generative AI, and to distinctly label AI-generated results. This requirement notably includes deepfakes, which can often be indistinguishable from genuine content.
Implementation Timeline and Penalties
Authorities have indicated that they do not intend to impose immediate penalties for violations of the new legislation. Instead, a one-year grace period will be provided, after which administrative fines will be enforced for breaches of the law. For instance, failing to label generative AI content could result in a fine of up to 30 million won (approximately 20,400 US dollars).
However, these amounts are relatively modest compared to potential fines in the EU, where non-compliance can incur penalties ranging from 1% of a company’s global turnover for minor infractions to 7% for violations related to the use of high-risk AI technologies.
Industry Concerns and Support Initiatives
Despite the progressive nature of these regulations, many founders and executives within South Korea’s tech sector have expressed frustration. Lim Chung-wook, co-head of Startup Alliance, noted dissatisfaction regarding the country’s role as a pioneer in this area. He highlighted concerns that some of the legislation’s language lacks precision, which may compel companies to adopt safer yet less innovative strategies to mitigate risks.
In response to these concerns, South Korean President Lee Jae-myung urged policymakers to listen to industry representatives and provide adequate support for startups and venture companies. He emphasized the importance of maximizing the industry’s potential through institutional support while preventing anticipated adverse effects.
The Ministry of Science and ICT is planning to establish a platform with recommendations and a dedicated support center for companies during the transitional phase. Additionally, they are considering extending the grace period if necessary to accommodate industry needs.
