As artificial intelligence advances, a growing number of young South Koreans are turning to generative AI for legal advice in relatively minor criminal cases.
The Maeil Business Newspaper reported Wednesday that a 23-year-old college student used generative AI to develop a legal strategy before being questioned by police over sexually insulting remarks made while playing an online video game.
“I explained the facts and trained the AI on relevant legal provisions and precedents. It helped me get cleared of charges,” the student was quoted as saying.
The report said college students are increasingly recommending AI tools to one another to prepare for police questioning, including conducting mock interviews or drafting possible responses. Users on online forums are also sharing prompts designed to generate the most favorable legal guidance.
Much of the use appears concentrated in minor disputes, such as traffic violations or online conflicts, which are often settled without a full trial.
The growing reliance on AI in legal contexts is not limited to South Korea. As generative AI systems become more sophisticated, their potential use in legal proceedings is being explored in several countries.
Legal experts, however, warn that AI-generated responses may cite fabricated cases or legal precedents. So-called “hallucinations” — responses containing false or misleading information — remain a persistent issue in current AI systems.
Matthew Lee, a barrister at Doughty Street Chambers in the United Kingdom, has warned about the risk of AI errors entering legal proceedings.
“Given that AI usage within the legal profession will continue to grow, heightened vigilance is essential to safeguard the integrity of the justice system,” Lee wrote in a September article for Counsel magazine, published by the Bar of England and Wales.
minsikyoon@heraldcorp.com
