Thailand’s new labeling requirements for medical devices, which include for the first time a unique device identification (UDI) requirement for software as a medical device (SaMD), take effect on June 20, 2026. The Notification of the Ministry of Public Health regarding Criteria, Methods, and Conditions on Labeling and Instructions for Use for Medical Devices 2025, which replaces a similar notification from 2020, was published in the Government Gazette on December 22, 2025. To ensure clarity, modernity, and patient safety, the regulation requires domestic manufacturers and importers to provide labels and instructions for use (IFU) that are clearly legible, complete, and free of false or misleading claims. It also permits IFU to be provided in electronic format, such as via QR codes, websites, or other digital channels—directly relevant to SaMD, where physical labels are impractical and electronic presentation is the natural medium. The notification distinguishes two categories for labeling language. Home-use medical devices (for lay users outside healthcare facilities) must have labels and IFU in Thai. Professional-use medical devices may display labels and documentation in either Thai or English. This distinction is significant for SaMD developers: software intended for clinical professionals may use English-language interfaces and IFU, while consumer-facing health applications must provide Thai-language content. Labeling and UDI Requirements Labels and IFU must include, at a minimum: Product name and intended purpose Quantity or volume Name and address of domestic manufacturer or importer Thai FDA approval number Lot, version, or serial number Manufacturing date and expiry date For SaMD, the version number requirement is particularly relevant. The regulation also mandates display of a UDI code for SaMD in risk category 2 (moderate-risk), category 3 (moderate- to high-risk), and category 4 (high-risk), according to Thailand’s medical device risk classification system (which complies with the ASEAN Medical Device Directive and the EU

The president of Thailand’s Supreme Court has issued new recommendations providing courts with criminal jurisdiction with a comprehensive framework for identifying and dismissing criminal cases brought in bad faith. Published in the Government Gazette on May 29, 2026, after being signed on May 25, the Recommendations of the President of the Supreme Court Concerning Bad-Faith Litigation in Criminal Cases B.E. 2569 were issued under Section 5 of the Act on the Organization of Courts of Justice. The recommendations took effect upon publication and represent a significant step in Thailand’s efforts to curb abusive criminal litigation, including strategic lawsuits against public participation (SLAPP). Background Section 161/1 of Thailand’s Criminal Procedure Code empowers courts to dismiss criminal cases filed dishonestly or with the intent to harass or take unfair advantage of a defendant. The new recommendations provide detailed guidance that courts previously lacked on identifying and handling such prosecutions. Definition of Bad-Faith Litigation Under recommendation 1, filing a criminal case in bad faith is defined broadly to encompass three categories: Harassment-type filings involving intimidation, threats, or creating unreasonable hardship for the defendant; Coercive filings designed to pressure the defendant into acting or refraining from acting for illegitimate benefit; and False or misleading filings that deliberately assert incorrect material facts or conceal such facts. Circumstances Indicating Bad Faith Recommendation 2 sets out specific circumstances that should raise a court’s suspicion that a filing may violate section 161/1. These include: Filing in a distant court far from the defendant’s domicile without benefiting the adjudication; Retaliation against the defendant’s advocacy for human rights, environmental protection, consumer rights, labor rights, or other public interests—effectively establishing an express anti-SLAPP framework; Retaliation against whistleblowers who disclosed corruption or unlawful conduct; Retaliation against individuals responsible for investigating the plaintiff’s wrongdoing or who concluded such an investigation; Filing multiple

Since the implementation of the Trademark Law 2019 on April 1, 2023, Myanmar has operated under a modern first-to-file trademark system that brings its registration framework closer to international practice. As the new regime continues to develop in practice, applicants are increasingly required to navigate formal examination requirements, substantive objections, and procedural deadlines with greater precision. This article provides a high-level review of the trademark examination process in Myanmar, focusing on the principal stages from initial review to approval, the types of objections commonly raised by the Intellectual Property Department (IPD), and the key considerations for responding effectively. A clear understanding of these issues is essential for applicants seeking to secure registration efficiently and to mitigate avoidable delays or refusals. Examination Process: Key Stages Trademark applications filed with the IPD undergo two stages of review. Formality Examination The IPD first verifies compliance with procedural requirements, including: Correct Nice Classification Clear mark representation Accurate applicant details Clearly defined goods or services Representative details, if the application is filed by a representative Other formality requirements cover translation and transliteration of any non-English or non-Myanmar elements in the mark, color claim details, applicable disclaimers, and payment of official fees. Deficiencies result in an office action requiring correction within 30 days, which may be extended upon request. Registrability Examination The IPD also assesses registrability. A mark may be refused if it: Lacks distinctiveness Is descriptive or generic Misleads the public or violates public order/morality Contains prohibited state symbols Only compliant applications proceed to publication. Responding to Office Actions Applicants must respond within 30 days of notification from the IPD. Depending on the nature of the objection, strategies may include submitting legal arguments for distinctiveness, providing evidence of acquired distinctiveness, filing appropriate disclaimers, clarifying descriptions such as color claims, or amending the listed goods

The surge in AI development has led to a desperate demand for large, high-quality training data. However, real-world data can be expensive to collect, difficult to access, and often subject to strict privacy and regulatory constraints. Synthetic data, which consists of artificially generated records that replicate the statistical properties of real-world data without reproducing specific individuals’ information, provides an appealing solution by generating artificial datasets at scale without relying on identifiable personal information. It combines speed, cost efficiency, and regulatory compliance, making it a sensible alternative for organizations seeking to reduce risks while maintaining data utility. When properly anonymized, synthetic datasets may fall outside the scope of laws such as the EU’s General Data Protection Regulation (GDPR) or Thailand’s Personal Data Protection Act (PDPA), reducing compliance burdens while still supporting high-quality model training. However, relying on synthetic data without rigorous legal due diligence could be a strategic mistake. It replaces one set of known risks (scraping, direct privacy liability) with a new set of complex liabilities. The narrative that synthetic data is a “silver bullet” for privacy and IP compliance is dangerous and could be misleading. While synthetic data addresses data scarcity, it also introduces new legal uncertainties. Legal counsel should anticipate downstream risks arising from compromised data sources. Models trained on unlawfully obtained data may need to be decommissioned, even if their outputs appear lawful. What is synthetic data? Synthetic data refers to artificially generated information created using AI techniques such as deep learning and generative models. Instead of copying real records, it reproduces the statistical patterns and relationships found in the original dataset. Synthetic data generally falls into three categories: Fully synthetic data – Entirely new data points generated from learned patterns. The model studies the structure of the original data and produces records that resemble real-world