Thailand
Tools
Tasks
Users
Scope
Training
Regulation
Insights
Information uploaded as at June 2026
AT A GLANCE
Thailand is expanding AI use primarily in policing, with tools such as AI-enabled robots (“AI Police Cyborg 1.0”), facial recognition, and smart CCTV supporting suspect identification, crowd monitoring, traffic enforcement, and fraud detection. A centralised AI platform to combat online scams is set to launch in 2026, and authorities are developing AI-supported social media monitoring for cybercrime and narcotics enforcement. Prosecutors are exploring AI for efficiency—using tools like ChatAI for document summarisation, drafting, and research—while judicial use remains limited within a regulated framework. Defence lawyers increasingly use AI for administrative support and legal research. Training initiatives are underway across police, prosecutors, and judges, with an emphasis on responsible and ethical AI use.
Thailand has no comprehensive statutory framework governing AI use in criminal proceedings. Draft AI Law Principles remains pending enactment, while the judiciary has filled the gap with its own guidance: the President of the Supreme Court's non-binding Recommendations (2025) apply to judges across all courts, including criminal matters, and the Civil Court's binding Regulations on AI-generated pleadings (2025) impose disclosure, verification, and certification requirements enforceable as contempt of court. General laws — including the Personal Data Protection Act, Computer Crime Act, and Criminal Code provisions on false evidence and document tampering — apply indirectly to AI misuse, but no legislation, rule, or case law directly addresses AI-generated evidence in criminal proceedings, and no relevant cases have been reported to date.
USE
Law enforcement
Operational support
As at June 2026, it has been reported that the Royal Thai Police Provincial Police Region 2, in cooperation with King Mongkut’s Institute of Technology Ladkrabang and Amata Corporation, is developing a robot powered by AI to support the work of inquiry officials (police officers) and to mitigate the issue of lack of resources with a potential to replace inquiry officials in the future.
Thailand is strengthening its response to online fraud by developing a centralised, AI-powered data management system for the Anti-Online Scam Operation Center (AOC). Scheduled to launch in 2026 with a 200 million baht budget, the system will connect multiple government agencies and allow them to share data in real time to more quickly detect, track, and stop online scams such as phishing, fake calls, and mule accounts.
Thailand’s Central Investigation Bureau (CIB), a specialised agency under the Royal Thai Police, has also announced that an IJUSTICE Chabot is being tested, to assist with interviews of sexual assault victims using psychological principles. An AI Carpool is also in the testing phase, and is a system to arrange for immediate transportation of the police officers, by selecting the most suitable vehicle for each mission. The AI analyses operational requirements, vehicle capabilities, availability, and specific conditions of each mission to recommend and provide the most suitable vehicle for CIB officers, ensuring efficiency, effectiveness, and optimal resource utilisation in every operation.
Data review and analysis
The Royal Thai Police have deployed AI-enabled robots such as “AI Police Cyborg 1.0”. These robots are equipped with 360-degree smart cameras, facial recognition, and AI video analytics that process live CCTV and drone footage to:
- Automatically detect and alert officers about wanted or high-risk individuals;
- Identify suspects;
- Support crowd monitoring and situational awareness;
- Help identify potential threats in real-time.
The data gathered by the AI-enabled robots is streamed to police command centers to assist coordination and public safety management. In May 2025, it was reported that AI Police Cyborg 1.0 had contributed to the capture of 14 fugitives. Deployment of AI Police Cyborg 1.0 has improved operational efficiencies by reducing the number of police officers having to be deployed for public events and is aimed to be a test for further development and usage.
Law enforcement and city authorities in Bangkok are also using AI-enhanced CCTV to enforce laws more efficiently. For example, CCTV systems with AI analyse footage to identify motorists riding on sidewalks, automatically read license plates, and trigger enforcement actions, resulting in fines and summonses issued without manual review. Similar AI systems can assist with traffic violations, public order monitoring, and urban surveillance tasks by automating the detection and documentation of offenses that previously required human monitoring.
In other locations such as Pattaya, police have installed AI-powered cameras that flag wanted suspects connected to large fraud networks. The system automatically alerts authorities, leading to arrests. In 2025, such cameras were used to arrest three key suspects involved in a major “mule account” fraud network.
It has been reported that the Ministry of Digital Economy and Society has agreed to an arrangement with: (1) the Technology Crime Suppression Division of the Central Investigation Bureau under the Royal Thai Police and (2) the Office of the Narcototics Control Board under the Ministry of Justice to set up a center to monitor social media activity with the aim of suppressing narcotic purchases via online platforms. The center will employ both physical resources and AI in its monitoring function to identify information to be sent to the Royal Thai Police and the Office of the Narcotics Control Board under the Ministry of Justice for further investigation and prosecution.
The collaboration between the RTP Provincial Police Region 2 and King Mongkut’s Institute of Technology Ladkrabang and Amata Cooperation, mentioned above, is also aiming to develop smart CCTV for surveillance in industrial areas, as well as forensic technology powered by AI.
The Central Investigation Bureau is testing a Detective Assistant chatbot to assist in the gathering and organisation of information obtained during investigations. It is also testing a tool named ENVI Invest, a template prompt to be used as a system to support the determination of issues to be investigated.
Prosecutors
Thailand’s Office of the Attorney General has engaged with AI technology research and experts to explore how it can support prosecutorial activities. Representatives from the Office and national research institutes have met to discuss applying AI in tasks such as case research, document summarisation, and evidence analysis to improve efficiency.
AI is an essential thing and a tool to assist in the workflow. It is promoted for public prosecutors to use AI in researching for various informing which will reduce the working process.
Case management
Training and workshops have been held in Thailand, where government technology agencies taught the prosecutor’s office to use AI tools like ChatAI. ChatAI is a generative AI tool developed by Thailand’s National Science and Technology Development Agency (NSTDA), to improve efficiency in internal work, such as summarising documents; drafting routine text; answering internal questions, and analysing data quickly.
Legal research, analysis and drafting support
As set out above, prosecutors in Thailand have been trained to use ChatAI to draft routine text.
Courts
Judicial use of AI remains limited, and is heavily circumscribed by Thailand’s regulatory framework (see below).
Case management
Thailand’s judiciary is exploring the potential of AI in case management and court support. Thailand’s Administrative Court has organised discussions with the Singaporean judiciary on AI applications in justice systems.
Thai institutions like the National Electronics and Computer Technology Center and the judiciary are also exploring the potential for 'smart court' systems that use AI for text and speech processing, translation, and chatbots or help systems for public access.
Defence
Administrative support
AI is increasingly used in Thai legal practice for non-substantive work such as managing appointments and schedules, tracking deadlines and court dates, automating repetitive form-filing and template documents, and client communication via chatbots or automated status updates.
Legal research, analysis and drafting support
There is increasing interest in AI systems used to support legal research and retrieve relevant legal information. Funding has been approved by the National Innovation Agency for an AI-powered legal research platform intended to serve as a ‘digital research assistant’ that can:
- Conduct comprehensive searches of legal provisions;
- Retrieve Supreme Court judgments;
- Identify relevant case law precedents with precision and speed.
The project is ‘designed to prepare the next generation of lawyers to leverage digital tools for enhanced professional performance’.
Victims
Under Thai criminal procedure law, injured persons (or victims) have independent standing in criminal proceedings. The Criminal Procedure Code recognises an injured person as a party with procedural rights separate from those of the public prosecutor, including the right to initiate a criminal proceeding under Section 28(2), and the right to participate as a co-plaintiff in a proceeding brought by the public prosecutor under Section 30, without the need to rely on any authority.
As at June 2026, there are no reported uses of AI by victims in criminal proceedings in Thailand. However, on 20 June 2026, the CIB launched CIBbot, an AI-powered legal chatbot developed by the CIB in collaboration with the National Research Council of Thailand (NRCT) and the National Institute of Development Administration (NIDA). The chatbot is currently in the phase of public testing, providing preliminary legal guidance on criminal complaints, legal procedures, and citizens' rights, aiming to improve public access to legal information and enhance the efficiency of legal services.
Training
The Royal Thai Police are investing in training programmes focused on AI use in investigations. As at July 2026, two intensive training camps had been offered to investigation officers, teaching them how to apply AI tools and analytics in real criminal investigations. The programmes covered AI for automation tasks, the analysis of large datasets, and streamlining document and evidence processing.
Prosecutors have received tool-specific training, such as for Thailand’s ChatAI tool (as discussed above).
Thailand has also hosted regional AI and rule-of-law training for judges. UNESCO and the UN Development Programme hosted judicial workshops in Bangkok in November 2025, organised with the Thailand Institute of Justice, where judges from across the Asia-Pacific region discussed ethical and responsible AI use in justice systems, focusing on fairness, transparency, and human rights. Key training elements included assessing AI's impact on justice systems (e.g., court administration, legal research, case management), addressing risks including algorithmic bias, lack of transparency, and erosion of judicial independence, and implementing safeguards such as explainability, human oversight, bias audits, and ethical standards aligned with human rights and UNESCO's Global Toolkit on AI and the Rule of Law. Participants explored real-world scenarios for responsible AI deployment to enhance access to justice while upholding fairness and public trust, with the Thailand Institute of Justice committing to adapt the toolkit for ongoing judicial training in Thailand.
As at June 2026, there is no dedicated training on deepfakes in Thailand.
Regulation
As at June 2026, there is no single, comprehensive regulation governing the use of AI in any judicial proceedings. However, there are related laws that indirectly address aspects of AI deployment, including the Personal Data Protection Act (2019) for data privacy in AI processing, the Computer Crime Act (2007) for cybersecurity risks in AI systems, and court-specific regulations on AI-assisted pleading drafting.
On 9 October 2025, a spokesperson for the Court of Justice commented that the use of generative AI in court proceedings must be governed by a clear framework and rules. The spokesperson referred to recommendations made by international organisations such as the Council of Europe, UNESCO, and OECD, including that AI usage must be within the framework of fairness, responsibility, transparency and disclosure (which appears to be a reference to the UNESCO Guidelines for the Use of AI Systems in Courts and Tribunals (2025)).
AI Regulations
Thailand currently has no statutory provisions that expressly regulate AI deployment within criminal proceedings in the Criminal Court.
The Draft AI Law Principles, developed by the Electronic Transactions Development Agency working with the Ministry of Digital Economy and Society with public consultations concluded in June 2025, marks Thailand's first draft of a comprehensive AI governance framework. The draft AI Law Principles represents a development and consolidation of the two-prior draft pieces of legislation relating to AI: (i) the Draft Royal Decree on Artificial Intelligence System Service Business, and (ii) the Draft Act on the Promotion and Support for National AI Innovation, which were introduced and proposed in 2022 and 2023, respectively (but did not capture the fast-growing development of AI technologies, in particular, generative AI).
Importantly, the current draft AI Law Principles have the objective of aligning the regulatory framework with advancements in AI developments, as well as relevant foreign laws and guidelines on AI technologies, in Thailand’s domestic context. In essence, the draft AI Law Principles cover a broad legal spectrum, including basic principles of liability in the use of AI, misuse, risk management by the sectoral regulator, regulator committee and expertise committee, legal representative of overseas service provider, notice of serious incidents, duty of care, text and data mining, sandbox, and AI governance clinic.
Under the Principles, AI systems are classified by risk profile, and providers and developers of AI must prepare evidence for this classification. All entities developing or deploying AI systems must ensure that they do not develop or deploy AI systems that are classified as prohibited risk under the Principles, and establish lawful bases for AI training data. They must also implement a formal risk management framework, and register or notify identified high-risk AI systems with the designated authority (the Electronic Transactions Development Agency is described as the coordinating regulatory authority for digital services).
Guidelines for practitioners
Though Thailand has referred to the UNESCO Guidelines for the Use of AI Systems in Courts and Tribunals, as at June 2026, Thailand has not formally adopted the Guidelines.
Recommendations of the President of the Supreme Court regarding the Use of AI Technology in Case Operations (2025)
The President of the Supreme Court has issued Recommendations on the Use of AI Technology in Case Operations, which outline a series of guidelines and an ethical framework for the integration of AI within the Thai judiciary. The Recommendations were issued on 19 September 2025 and apply to a broad range of individuals within the justice system, including judges and court personnel working within the court who are involved in case-related functions and may use AI in the course of their duties. The Recommendations are non-binding, and as at June 2026, there have been no publicly reported sanctions in cases of non-compliance.
|
Foundational Mandates and Judicial Independence |
The use of AI technology must strictly adhere to the fundamental principles of the administration of justice, ensuring that the administration of justice remains fair, impartial, and free from bias. Crucially, judicial independence must be maintained in all legal proceedings. AI is categorised strictly as a support tool for operational efficiency, and judges are prohibited from allowing AI technology to influence or interfere with their exercise of discretion in the determination of cases. |
|
Standards of Responsibility and Risk Mitigation |
Judges bear the highest degree of responsibility to uphold the integrity of the judicial process, which entails an affirmative duty to remain cognisant of AI's inherent limitations. These include algorithmic bias and system processing constraints, the potential for incomplete, inaccurate, or unreliable data outputs, and the essential requirement to verify the accuracy and completeness of all AI-generated results before use. |
|
Confidentiality and Data Protection Protocols |
AI integration must not compromise judicial information security. The Recommendations of the President of the Supreme Court impose strict prohibitions, including the prohibition on disclosing official confidential information through unauthorised AI access and submitting or processing personal data from case files via AI systems (Article 2). |
|
Authorised and Prohibited Scopes of Application |
The Recommendations of the President of the Supreme Court distinguish between permissible administrative support and prohibited core adjudication functions. AI use is allowed for administrative or academic tasks, such as information retrieval, mathematical calculations, data categorisation, case management planning, and drafting preliminary materials or documents (such as summarising facts relevant to rulings). The use of AI as a legal research tool is permitted under Article 5 of the Recommendations, but users must verify the accuracy of any information obtained through such tools. However, judges are strictly forbidden from using AI for assessing, predicting, or analysing the probability of disputed issues to determine a case. |
|
Transparency and Institutional Oversight |
Where AI-generated information is used in case-related duties, such use must be appropriately disclosed in accordance with applicable rules and confidentiality obligations. Where AI has been used, the user must disclose such use. The Office of the Judiciary is mandated to select and provide reliable AI technologies that facilitate access to justice while ensuring the protection of personal data and official confidential information. |
Civil Court Regulations on the Use of AI Technology in the Preparation of Pleadings or Other Documents for Submission to the Court B.E. 2568 (2025)
The Civil Court Regulations on the Use of AI Technology in the Preparation of Pleadings or Other Documents for Submission to the Court B.E. 2568 (2025) impose a number of legal obligations on parties and lawyers. The Regulations apply to parties and lawyers who use AI to prepare pleadings or include AI-generated content in documents submitted to the Civil Court.The Regulations are binding and enforceable, and parties and lawyers remain fully responsible for AI-generated content. Any non-compliances may result in sanctions imposed by the relevant official and may be deemed to be contempt of court under Section 31 of the Civil Procedure Code.
|
Purpose and Scope of Regulations |
In recognition of technological advancements, the Regulations of the Civil Court were announced to ensure the fairness and integrity of judicial proceedings and to maintain public confidence in the administration of justice. The Regulations of the Civil Court apply whenever parties and lawyers use AI to prepare pleadings or incorporate AI-generated outputs into court submissions. |
|
Professional and Ethical Responsibility |
Lawyers using AI technology are mandated to maintain the highest standards of honesty and professional responsibility toward the judicial process. All usage of such technology must comply with applicable laws, lawyers’ professional standards, and the code of ethics governing the legal profession. |
|
Mandatory Verification and Accuracy |
Before submitting any documents prepared by AI to the Court, the party and lawyer have an affirmative duty to ensure that all facts, statutory provisions, and cited evidence are accurate and complete in all respects. The party and lawyer must verify that the content does not contain distorted or false information, or biased data. |
|
Strict Disclosure and Labeling Requirements |
For the purpose of transparency, any use of AI must be explicitly disclosed within the document using specific formatting. The portion of content derived from AI must begin with the statement: ‘[t]he following content has been generated using AI technology’, and the AI-generated section must end with ‘[e]nd of content generated using AI technology.’ At the end of any pleadings, the party and lawyer must include a formal certification stating that: ‘Certain portions of this (title of document) have been prepared using AI technology, and I have reviewed and certify the accuracy of the facts and legal provisions prior to submission to the Court.’ |
|
Legal Liability and Accountability |
The party and the lawyer submitting the document bears full legal and ethical responsibility for all AI-derived content as if it had been drafted personally. Specifically, no party or lawyer may rely on technical errors of the AI as a grounds for avoiding responsibility or liability. |
|
Enforcement and Sanctions |
Non-compliance with Regulations of the Civil Court by a party or the lawyer may be a contempt of court under Section 31 (1) of the Civil Procedure Code, which is punishable by a term of imprisonment not exceeding 6 months and/or a fine of not exceeding five hundred Baht under Section 33 of the Civil Procedure Code. Judges or authorised officials are required to report violations of Regulations of the Civil Court to their superiors for further consideration and order. |
The Bangkok South Civil Court has also issued similar AI regulations, which were effective on 28 May 2026. These regulations have substantially the same scope as those of the Civil Court, applying to parties and lawyers who use AI to prepare pleadings or incorporate AI-generated content into court submissions. They require disclosure of AI use and verification of the accuracy of AI-generated content before filing. However, unlike the Civil Court Regulations, the Bangkok South Civil Court Regulations do not expressly provide specific sanctions for non-compliance.
Ongoing initiatives to prepare guidelines for lawyers
The Lawyers Council of Thailand Under the Royal Patronage is the statutory professional body responsible for regulating and supervising the lawyers in Thailand including licensing lawyers and setting ethical and professional standards. In addition, the Lawyers Council also performs an important public-interest function by promoting access to justice, including through legal aid, free legal advice, and volunteer lawyer services at police stations and courts nationwide. It is reported that the Lawyers Council of Thailand are working with the Faculty of Law, Thammasat University in consultation with the Electronic Transactions Development Agency to prepare practical guidelines for the use of AI by lawyers and legal advisors.
Criminal procedure rules
There are currently no specific rules governing the use of AI in Thai Criminal Procedure Code (unofficial English translation updated 2002). Currently, no amendment has been made or proposed to the Criminal Procedure Code to address recent developments relating to AI usage.
However, general admissibility principles under Criminal Procedure Code, Sections 226-231 and Electronic Transactions Act B.E. 2544 Section 11 of non-denial of electronic evidence admissibility, apply equally to AI-generated or assisted evidence like deepfakes.
The Recommendation of the President of the Supreme Court, discussed above, applies to judges across all courts, including the Criminal Court. Accordingly, AI may be used by judges in criminal proceedings, subject to the limitations and safeguards prescribed under the Recommendation of the President of the Supreme Court.
Deepfakes and synthetic media
If evidence submitted in a criminal case is created using AI deepfakes, it may constitute a criminal offence under Section 180 of the Criminal Code, which is punishable by an imprisonment of not more than 7 years and a fine of not exceeding 140,000 Baht (the punishment is imprisonment of not more than 3 years or a fine of not more than 60,000 Baht or both if such evidence is used in a civil case). Offences under Section 180 concern the adducing and producing of false evidence in any judicial proceeding.
Use of AI deepfakes to create false evidence to convince an inquiry officer that a criminal offence has been committed, or that a more severe criminal offence has been committed, may constitute a criminal offence under Section 179 of the Criminal Code. This is punishable by a term not exceeding two years imprisonment, or a fine of no more than 40,000 Baht or both.
Section 264 of the Criminal Code similarly stipulates penalties for document tampering, covering forgery, alteration, or fake signatures or seals intended to deceive and cause harm.
Moreover, Section 16 of the Computer Crime Act B.E. 2560 (2007), as amended in 2017, criminalises the importation into publicly accessible computer systems of harmful digital content, including the creation or dissemination of obscene computer data such as pornographic material.
The Draft AI Law Principles (discussed above) include a principle to criminalise the use of AI deepfakes for pornographic purposes or to interfere with an election. However, the Principles are of no binding force pending formal legislative enactment.
Data Protection Legislation
Given the lack of specific AI legislation in Thailand, collection of personal data such as names, biometric identifiers, or financial records by an AI system for training, analysis and decision making purposes must strictly comply with the Personal Data Protection Act 2019, Thailand’s personal data protection law. The Act entered into force on 1 June 2022. Although the Personal Data Protection Act 2019 does not expressly refer to AI or AI systems as it was enacted before the widespread use of AI in Thailand, it imposes obligations including lawful basis for processing data (e.g., consent or legitimate interest), data maximisation, security safeguards, transparency via privacy notices, and restrictions on cross-border transfers, all of which apply whenever AI handles personal data of individuals in Thailand. In particular, the collection of personal data under the Act must be for a lawful purpose and be directly relevant to, and necessary for, the activities of the Data Controller, who must inform the data subject of the purpose of the collection for use or disclosure of the personal data, and the personal data to be collected and the period for which it will be retained.
Compliance under the Personal Data Protection Act 2019 arises through general data processing rules, with the Personal Data Protection Committee issuing soft guidance on automated decision-making and profiling transparency.
Risk-based approaches, such as anonymisation before AI training, consent verification, and processing activity records, may serve to prevent violations of the Personal Data Protection Act, pending Thailand's forthcoming AI legislation to be drafted based on public opinions on the draft AI Law Principles which is expected to introduce explicit high-risk AI registration and sandbox requirements.
Cybersecurity Laws
The Computer Crime Act B.E. 2550 (2007), as amended in 2017, does not contain specific provisions directly addressing AI, however, it imposes penalties for computer-related offences that may relate to the misuse of AI, such as cases involving the creation or dissemination of unlawful content via AI.
State authorities may pursue prosecutions under this Act for offences including pornography or fabricated materials generated by AI (e.g., generative AI or deepfakes), specifically under Sections 14(4) and 16 of the Computer Crime Act that criminalises the importation into publicly accessible computer systems of harmful digital content.
- Section 14(4) criminalises the creation or dissemination of obscene computer data, such as pornographic material, to be punishable by imprisonment for a term not exceeding five years, or a fine not exceeding THB 100,000, or both.
- Section 16 criminalises fabricated, edited, or electronically altered photographs of individuals that impair reputation, incite public hatred or contempt, or occasion shame (encompassing AI-generated deepfakes) to be punishable by imprisonment for a term not exceeding three years, a fine not exceeding THB 60,000, or both.
Human rights
Thailand has not enacted statutory provisions mandating law enforcement or judicial authorities to conduct human rights impact assessments for AI deployments. Current protections arise from the Constitution of the Kingdom of Thailand, B.E. 2560 (2017), which safeguards equality before the law, personal privacy, and due process rights under Section 27. The Constitution of the Kingdom of Thailand also provides the right to personal privacy under Section 32, which was subsequently supplemented by the Personal Data Protection Act B.E. 2562, and right to a fair trial is provided under Section 68 of the Constitution of the Kingdom of Thailand. In addition, in the international degree, the Convention of the Rights of the Child and the International Covenant on Civil and Political Rights also provides protections to the right to a fair trial and the right to privacy under Article 16 and 40, and Article 14 and 17, respectively.
In terms of ongoing initiatives, the Draft AI Law Principles (discussed above) recommend risk-based evaluations for high-risk AI systems with potential implications for fundamental rights, including where AI systems for biometric identification and analysis for law enforcement or border control, or where AI systems are deployed in the administration of justice. The Recommendation of the President of the Supreme Court (discussed above) promotes awareness of the limitations that may arise from algorithmic biases which may prejudice the rights and liberties of the Thai people to have their cases adjudicated by judges in accordance with the Constitution and laws (Article 1).
Outlook
As at June 2026, no statutory provisions expressly regulate the deployment of AI within criminal proceedings. The Draft AI Law Principles are expected, once implemented, to constitute Thailand's comprehensive AI governance framework. The framework prioritises risk-based oversight of high-risk AI applications with potential implications for fundamental rights and public safety. Although not expressly directed at criminal justice administration, its foundation holds prospective applicability as a regulatory prototype for subsequent sector-specific enactments.
Thailand’s National AI Strategy and Action Plan, which sets out strategic steps to be implemented by relevant governmental agencies from 2022 to 2027 was approved by the Cabinet on 26 July 2022, signalling Thailand’s readiness to the new era of AI technology. The Strategy and Action Plan is aimed at creating social impact by using AI to enhance national security and safety, with one of the key strategies of the Strategy and Action Plan being to promote the use of AI in public sectors.
In February 2025, Thailand also engaged in AI cooperation through non-binding formats such as Memorandums of Understanding with China and UNESCO.
CASES
As at June 2026, there have been no reported cases in Thailand on AI use, or the admissibility of AI-generated evidence.