South Korea Deploys AI-Powered Surveillance System to Combat Crypto Market Manipulation
The Financial Supervisory Service (FSS) in South Korea has rolled out an AI-powered market surveillance system to monitor and detect unfair trading practices across the country's cryptocurrency markets. The platform combines generative AI with machine-learning algorithms to track suspicious activity, including price manipulation schemes known as 'racehorse' and 'cage'.
The system is designed to overcome resource constraints faced by human investigators, who must monitor thousands of digital assets trading 24/7 across multiple exchanges. It pairs Benford's law with machine-learning models to flag wash trading and coordinated group trading, and automatically scans related news coverage and official exchange announcements for signs of manipulation.
When suspicious activity is detected, the FSS requests granular trade data directly from the affected exchange. The platform also monitors digital channels such as YouTube, online message boards, and private messaging app chat rooms to identify potential front-running or false information dissemination. However, human oversight remains central to the process, with FSS investigators evaluating AI-generated reports before deciding whether to launch formal enforcement proceedings.
Industry observers have expressed skepticism about relying on AI for regulatory enforcement, citing concerns about data accuracy and the risk of false positives. Markus Levin, co-founder of XYO, questioned how regulators can ensure these systems operate within clearly defined operational boundaries.