You are using an outdated browser and your browsing experience will not be optimal. Please update to the latest version of Microsoft Edge, Google Chrome or Mozilla Firefox. Install Microsoft Edge

April 2, 2026

Thailand Insurance Industry: AI and Privacy Regulatory Updates

Thailand’s Personal Data Protection Act (PDPA) enforcement has entered a new phase, and the insurance industry is squarely in the regulatory spotlight. The Personal Data Protection Committee (PDPC) considers insurers “large-scale” processors of sensitive data—including health records, financial information, and biometric data—making the sector a focal point for enforcement action. In August 2025 alone, the PDPC issued administrative fines totaling THB 21.5 million, and fines for individual violations have ranged from THB 50,000 to THB 2 million. The PDPC has also deployed its “Eagle Eye Crawler,” an AI-driven surveillance tool that monitors websites around the clock for data leaks and noncompliant privacy notices. This article highlights the key regulatory developments directly affecting insurers and outlines practical steps toward compliance.

What Has Changed: OIC and PDPC Alignment

The Office of Insurance Commission (OIC) has synchronized its sector-specific rules with the PDPA through the Notification on Customer Personal Data Protection (No. 2) B.E. 2568 (2025). The combined effect of the PDPC’s general enforcement push and the OIC’s sectoral guidance creates four critical compliance areas for insurers.

  • Consent unbundling. Consent for marketing must be strictly separated from the core insurance contract; bundling marketing consent into the policy application is no longer permissible.
  • Agent and intermediary oversight. Insurance intermediaries are generally classified as data processors, meaning that insurers—as data controllers—must provide specific written instructions and security protocols to all agents and brokers. A 2026 enforcement trend shows controllers being held liable for the “weak security” of their vendors and downstream processors.
  • Enhanced privacy notices. Insurers must provide a summary privacy notice alongside the full policy, plainly stating categories of data, purposes, lawful bases, disclosure recipients, cross-border transfers, retention periods, data subject rights, and easy marketing opt-out channels.
  • DPO registration and ROPA. All organizations involved in “regular or systematic monitoring of data subjects on a large scale”—expressly including insurance—must appoint and register a data protection officer (DPO). The absence of a registered DPO or an outdated record of processing activities (ROPA) that fails to map agent-level data flows is now considered a high-risk compliance gap.

AI in Insurance: Draft PDPC Guidelines

The PDPC’s draft AI guidelines carry particular significance for insurers. The guidelines single out insurance risk assessments as an example of automated decision-making that produces legal effects or significantly affects data subjects. Organizations using AI-driven tools for underwriting, claims processing, or policy recommendations must implement a human-in-the-loop mechanism with actual authority to overturn AI decisions and must document processes for data subjects to request review. A data protection impact assessment (DPIA) is required for high-risk AI projects, including automated decision-making with legal effects and large-scale processing of sensitive data. Leakage of sensitive health or financial data through AI systems is categorized as high risk, requiring notification to both the PDPC and affected data subjects without delay.

Cross-Border Data Transfers

For multinational insurance groups, a binding corporate rules (BCRs) regulation became fully effective on February 17, 2026, providing a formal mechanism for intragroup cross-border transfers. Groups that already hold GDPR-approved BCRs may use a “fast-track” process by submitting their existing BCRs together with a Thailand addendum. Alternatively, Standard Contractual Clauses based on the ASEAN Model Contractual Clauses may be used for transfers to third-party reinsurers or service providers outside Thailand.

Practical Compliance Steps

Given the current regulatory landscape, insurers should consider the following immediate and near-term actions.

  • Governance and organization. Register a DPO with the PDPC if not already done, and ensure that the DPO has a direct reporting line to senior management with sufficient authority and resources to fulfill the role. Update the ROPA to comprehensively map all processing activities, including data flows through agents, brokers, and third-party administrators.
  • Consent architecture overhaul. Redesign application forms and digital onboarding flows so that marketing consent is presented as a separate, clearly labeled opt-in, entirely distinct from the consent required for the insurance contract itself. Ensure that refusal to consent to marketing does not affect the customer’s ability to obtain coverage.
  • Agent and vendor compliance program. Issue updated written instructions and security protocols to all insurance intermediaries classified as data processors. Review and strengthen data processing agreements with all third-party processors, including specific provisions for PDPA responsibilities, security standards, audit rights, breach notification obligations, and end-of-term data deletion or return. Implement a periodic audit cycle—rather than relying on static contractual commitments—to verify vendor compliance.
  • Privacy notice refresh. Prepare a concise summary privacy notice for distribution alongside insurance policies, covering all required elements under the OIC guidance. For digital tele-sales, implement prerecording disclosures informing customers that their voice or image data will be processed under the PDPA.
  • AI and automated decision-making readiness. Conduct DPIAs for all AI-driven underwriting, claims, and risk-assessment tools currently in use or under development. Establish a documented human-in-the-loop process for any automated decision that produces legal effects on policyholders, including a clear escalation path and a mechanism for data subjects to contest decisions.
  • Breach response preparedness. Ensure that internal incident response plans can meet the 72-hour notification deadline to the PDPC, with particular attention to AI-related data leakage scenarios.
  • Cross-border transfer mechanism. For multinational groups, evaluate whether BCR certification—including the fast-track route—or SCCs provide the most efficient path for data transfers to group entities or reinsurers abroad.

Outlook

Thailand’s insurance sector faces a significantly more demanding compliance environment as PDPA enforcement matures and OIC alignment tightens. The convergence of stricter consent rules, expanded liability for intermediary conduct, new AI governance expectations, and a workable cross-border transfer framework means that insurers must move from reactive compliance to proactive data governance. Organizations that address these areas systematically—beginning with DPO registration, ROPA updates, and consent architecture—will be best positioned to manage regulatory risk and maintain the trust of their policyholders.

RELATED INSIGHTS​ 

August 1, 2025
On July 30, 2025, Myanmar’s Cybersecurity Law No. 1/2025 came into effect with the State Administration Council’s issuance of Notification 113/2025. The law, which was enacted on January 1, 2025, aims to regulate various aspects of digital security and online activities. Below are some key provisions, implications, and penalties under the Cybersecurity Law. Extraterritorial penalties. The law contains an important provision that authorizes penalties against Myanmar citizens who are found guilty of violations, even if these occur outside the country’s borders. VPN definition and regulation. Virtual private networks (VPNs) are defined by this law as specific systems that function as backup networks by using technological means in order to ensure the safety of linking networks to each other. This definition sets the framework for subsequent regulations and penalties associated with VPN usage. The law does not restrict individuals or entities from using VPNs; it regulates VPN service providers. Penalties for unapproved VPN services. Establishing a VPN or providing VPN services without approval from the designated ministry (to be appointed later by the government) can result in significant penalties. For individuals, the punishment may be imprisonment for 1–6 months, a fine of MMK 1–10 million (approx. USD 476–4,760), or both, with the proceeds of the violation being confiscated. If the violator is a company or organization, the minimum fine will be MMK 10 million, and the proceeds will be confiscated. Government oversight. The ministry designated by the government is authorized to investigate and take control of cybersecurity services and digital platform services for national defense and security purposes, or upon request from a government department or organization in accordance with respective laws. Licensing requirements. The Cybersecurity Law introduces two types of licenses, valid for a period of 3–10 years, for (1) cybersecurity services and (2) digital platform providers. Digital platforms with
August 1, 2025
On July 21, 2025, Thailand’s National Cyber Security Agency (NCSA) released a draft amendment to the Cybersecurity Act B.E. 2562 (2019) for public hearing, aiming to address the rapid evolution of technology and increasing complexity of cyber threats. The proposed changes to the country’s cybersecurity framework would extend regulatory oversight to cloud service providers and data center operators hosting data for critical information infrastructure (CII) organizations regulated under the Cybersecurity Act. The NCSA will accept comments on the draft until August 5, 2025. Following the close of the public consultation period, the draft amendment will be subject to further revision during the legislative process. Key proposed amendments are discussed below. Expanded Critical Infrastructure Scope The Cybersecurity Act currently applies only to state agencies, supervising or regulating organizations, and designated CII organizations as announced by the National Cyber Security Committee (NCSC). It defines CII organizations as public or private organizations related to or providing national security, significant public services, banking and finance, information technologies, telecommunications, transportation and logistics, energy and public utilities, or public health. The draft amendment expands the scope of CII organizations to include public and private organizations related to or providing industrial work (to be further defined in subregulations) as well as service providers that store or possess data for CII organizations, such as cloud and data center service providers. CII organizations must comply with cyber threat reporting requirements and are subject to the NCSA’s interception powers. Updated Definitions and New Terminology The draft amendment more clearly distinguishes between “cyber threats” (which have yet to occur but have the potential of causing damage or impact) and “cyber incidents” (which have already occurred and have caused or are expected to cause damage or impact). The draft amendment also expands the definition of “cybersecurity” to explicitly cover both prevention
July 30, 2025
Artificial intelligence (AI) model training and data scraping are essential processes in the development of modern AI systems. AI model training involves using large datasets to teach machine learning algorithms to recognize patterns, make predictions, or generate new content. Data scraping refers to the automated extraction of information from websites or digital sources, often to assemble the vast datasets required for effective AI training. As these practices become more widespread, questions about the legality of using third-party content—especially copyrighted works—have become increasingly important. In Thailand, the legal landscape for AI developers is shaped primarily by the Copyright Act, which presents unique challenges due to the absence of a fair-use exception. This article examines the copyright-related risks and legal uncertainties facing AI developers under Thailand’s current copyright law and practices, offering strategic guidance for navigating this complex environment. Copyright Risks in AI Scraping and Training Thailand’s Copyright Act does not provide a broad fair use or fair dealing exception, unlike some other jurisdictions, such as the United States. This absence has significant consequences for AI developers: No general defense for AI training: Any use of copyrighted material for AI model training is presumed to be infringing unless a specific, narrow statutory exception applies or explicit permission is obtained from the rights holder. There is no general legal basis for using copyrighted works in AI training without authorization. Increased rights clearance burden: Developers must identify and secure licenses for every copyrighted work included in their training datasets. Given the scale and diversity of data required for effective AI models, this process can be both impractical and costly. Legal ambiguity and litigation risk: The lack of clear statutory guidance or case law leaves developers in a legal gray area. There is no established precedent clarifying whether certain uses of copyrighted material for
July 24, 2025
Thai authorities have escalated efforts to block unlawful cross-border digital asset business operators. On June 19, 2025, the Ministry of Digital Economy and Society (MDES) issued a notification empowering it to ban internet access to operations or services offered by digital asset business operators who lack licenses from the Thailand Securities and Exchange Commission (SEC) under the Emergency Decree on Digital Asset Businesses B.E. 2561 (2018). This ban, issued under the 2023 Royal Decree on Measures for the Prevention and Suppression of Technology Crime, particularly aims to block Thai users’ access to services offered by unlicensed offshore digital asset providers via their own apps or websites or through public social media platforms. Compliance Requirements The notification requires internet service providers and social media platforms selected by MDES to immediately impose internet access restrictions on identified apps, websites, and IP addresses of illegal operators upon receiving MDES orders. Takedown Orders There are two tracks for competent officials at MDES to issue orders to operators: If the competent official is notified by the SEC of licensing noncompliance by a particular digital asset business operator, the competent official can issue a takedown order to the operator upon approval from the permanent secretary of MDES. If the competent official independently discovers, or receives a complaint from any third party other than the SEC, that a digital asset business operator may have violated licensing requirements, the competent official can ask the SEC to verify and confirm the relevant facts and noncompliance before seeking approval from the permanent secretary of MDES to issue the takedown order. Streamlined Enforcement Prior to this notification, the SEC could obtain takedown orders only from Thai courts under the 2007 Computer Crime Act to take down or block access to unlicensed digital asset platforms and apps. This was a relatively