South Korea’s lawmakers and financial regulators are pushing for new legislation to address insurance fraud – and the urgency behind that push is being driven by documented cases of AI being used to defeat detection systems that existing laws were not designed to handle.
On September 16, the “National Assembly Forum on Preventing Insurance Crime” held its launch ceremony and inaugural seminar at the National Assembly Members’ Office Building in Yeouido, Seoul, according to The Asia Business Daily.
The forum was hosted by Assemblyman So Byung-hoon of the Democratic Party of Korea and Jeong Jeom-sik, floor leader of the People Power Party – cross-party involvement that signals the issue has moved from regulatory discussion into formal legislative process.
The scale of the problem frames why legislators are now involved. Insurance fraud payouts in South Korea reached a record 1.16 trillion won – approximately US$757 million – in 2025, up 6.9 billion won, or 0.6%, from the previous year, according to the Financial Supervisory Service (FSS). Fraudulent payouts have risen in consecutive years, from 1.08 trillion won in 2022 to 1.12 trillion won in 2023, the Korea Times reported.
Those figures only capture what is detected. The Financial Services Commission (FSC) estimates total exposure – including undetected cases – at around 9 trillion won, nearly eight times confirmed payouts, The Asia Business Daily reported. By product line, long-term non-life insurance accounts for 44.7% of detected fraud, followed by auto insurance at 22.4%, life insurance at 21.8%, and general non-life at 11.2%.
Regulators have been direct about the link to premiums. An FSC official stated that fraudulent payouts drive higher premiums, and that fraud touching health insurance benefits raises concerns about the long-term sustainability of the health insurance fund.
That sits within a broader regional cost environment. WTW’s 2026 Global Medical Trends report projects medical cost increases across Asia-Pacific will reach 14% in 2026, up from 13.2% in 2025 – the highest rate of any region globally. Fraud, waste, and abuse was cited by 38% of insurers surveyed as a driver of medical inflation.
Read next: Organized fraud is outgrowing Asia-Pacific’s cyber insurance market
At the forum’s seminar, Kwon Daeyoung, vice chairman of the FSC, identified two structural gaps driving the legislative push. “In order to respond quickly to insurance fraud that exploits artificial intelligence, it is crucial for relevant organizations to share information about suspected cases, and legislative measures, such as revising related laws, are also needed to support this,” Kwon said.
He also described how organised fraud networks have changed: “Insurance fraud has evolved beyond individual misconduct into an organized crime involving collusion among medical institutions, repair shops, brokers, and agents. Generative AI and deepfakes are now being misused for such fraud as well.”
The Korea Times reported that 2025 saw South Korea’s first prison sentence for AI-based insurance fraud – a man in his 20s was sentenced to two years for using AI to manipulate diagnoses and extend hospital stays, defrauding 11 insurers of 150 million won. A separate case involved a woman in her 40s who filed 39 claims over three years for conditions ranging from fractures to cancer; an insurer’s cross-check found that document formats, hospital seals, and doctors’ signatures had all been fabricated using AI.
An FSS official acknowledged the limits of existing tools: “Recent insurance fraud has become so sophisticated through the use of AI that it is difficult to detect not only with the naked eye but also with conventional screening systems.”
The challenge is compounded by gaps in data sharing and cross-verification. The FSC said Korea Credit Information Services, the Korea Insurance Development Institute, and individual insurers already operate AI-based insurance-fraud systems, but information sharing between institutions remains insufficient and cross-verification infrastructure is lacking. The government task force is examining ways to expand information sharing, including access to original documents for verification, and develop AI-based fraud-pattern analysis and risk indicators.
The forum brings together the National Assembly, government agencies, investigative bodies, research institutions, and the insurance industry. The General Insurance Association of Korea and the Korea Institute of Criminology and Justice are serving as organising bodies, with the association acting as secretariat.
Two research presentations at the September 16 seminar addressed the institutional gaps directly. Professor Jung Youngjin of Inha University’s Department of AI and Data Law outlined proposals for information-linkage systems between public and private insurers. Dr. Lee Bomi, a former researcher at the Korea Institute of Criminology and Justice, presented frameworks for inter-agency role allocation and joint investigation structures.
The forum’s outputs are intended to feed directly into National Assembly proceedings.
Lee Byungrae, chairman of the General Insurance Association, framed the challenge as one beyond what individual organisations can resolve: “Since it is difficult for individual organizations to fully address insurance crime on their own, it is necessary for the National Assembly, government, investigative agencies, research institutions, and the insurance industry to work together to establish a systematic, cross-ministerial response framework.”
Read next: NHIS phishing scam exposes gaps in South Korea fraud insurance
The forum is one of two parallel mechanisms now active. The FSC launched its "AI-Based Insurance Fraud Prevention System Task Force" in June and has set a September 2026 deadline for completing its platform development plan, with legal amendments and platform upgrades expected from October.
For insurance professionals with exposure to the Korean market, that October date is the point at which the regulatory picture becomes explicit. Changes to how fraud data is shared across public and private sectors are likely to affect claims investigation processes, documentation standards, and how intermediaries interact with fraud-reporting systems.
The Korea Times noted that industry insiders consider fully digital indemnity claims platforms – where medical data is transmitted directly from hospitals to insurers – the most viable near-term defence against document manipulation. That view, alongside the legislative and regulatory activity now under way, points to a claims environment in South Korea that is set to look materially different by early 2027.
