Last Updated on September 25, 2026 by Jeff Tomas
What is Thailand actually building to make AI part of its economy and public services? The effort reaches well beyond a single AI law or technology project: it combines a national strategy with plans for data and cloud infrastructure, Thai-language AI, workforce training, and rules for responsible use.
Some pieces are already underway, while proposed infrastructure and draft regulations are still taking shape. For context, Thailand’s National AI Strategy and Action Plan sets the broader direction for developing talent and expanding adoption. The details below are current as of September 2026, starting with the strategy behind the build-out.
Key Takeaways
- Thailand’s 2022-2027 National AI Strategy sets priorities for infrastructure, skills, innovation, adoption, and governance, with the official strategy overview outlining its national scope.
- New plans name a National AI Cloud, an upgraded LANTA II supercomputer, and a National AI Data Bank. Thailand is also developing Thai-language AI models and datasets.
- Workforce targets include training at least 160,000 people in big data and AI by 2027, alongside plans to certify 50,000 industrial workers and 5,000 AI specialists by 2030.
- Thailand’s draft AI Act proposes risk-based rules and transparency duties, but it isn’t law yet. See Thailand’s proposed AI regulatory framework.
Thailand’s AI Policy: The 2022-2027 plan and its targets
Thailand’s Cabinet approved the National AI Strategy and Action Plan 2022-2027 on July 26, 2022. Its vision is to build an effective AI ecosystem that supports economic growth and improves quality of life. The plan sets direction and measurable ambitions, but its targets should not be confused with results already achieved.
The strategy has five pillars: readiness in governance, ethics, law, and regulation; sustainable AI infrastructure; human capability and education; research and innovation; and AI adoption across public and private sectors. Together, they link the rules and resources needed to develop AI with its practical use in different industries.
Who is steering the national AI effort?
The National AI Committee provides national oversight and coordination. It helps keep government agencies aligned as they carry out the strategy, rather than assigning the entire effort to one ministry. The OECD’s profile of Thailand’s AI strategy also describes the plan as a coordinated policy initiative.
Two ministries led the strategy’s development. The Ministry of Higher Education, Science, Research and Innovation (MHESI) focuses on research, education, and building AI expertise. Meanwhile, the Ministry of Digital Economy and Society (MDES) leads digital policy and helps advance the conditions for wider AI use.
Other agencies carry specific parts of the work. The National Science and Technology Development Agency (NSTDA) contributes research and technical capacity, while the Office of the National Digital Economy and Society Commission supports digital-economy policy and coordination. NECTEC, part of NSTDA, also contributes technical expertise to the national AI effort. Its Cabinet announcement on the strategy outlines key ambitions.
What Thailand hopes to achieve by 2027
The plan sets targets for skills, innovation, adoption, and investment. It calls for more than 30,000 AI talents, at least 100 AI research and development prototypes, and 600 agencies using AI innovation. It also sets an ambition for annual digital infrastructure investment growth of 10%.
These figures are published goals, not proof that Thailand has met them. The strategy also seeks to grow AI-related economic value and improve social outcomes, including quality of life. In practice, that means developing useful applications, strengthening local research, and expanding AI use in both government and business.
The ambition is broad, but progress depends on the agencies translating national targets into funded programs, skilled teams, and working services. For a closer look at how private investment fits this policy direction, see Thailand’s enterprise AI buildout.
The infrastructure Thailand is building for AI
AI needs more than computing power. Public agencies also need secure ways to share usable data, while Thai-language models need local material to produce relevant results. Thailand’s plans connect these building blocks through government cloud services, shared data systems, and homegrown language-model work.
Government cloud and national data platforms
The Government Data Center and Cloud Service (GDCC) gives public agencies shared cloud infrastructure for hosting systems and services. By August 2026, it supported more than 4,000 government systems. Thailand has also described a shift toward a broader National Cloud, extending the cloud-first direction for public services. The World Bank’s Thailand Digital Data Infrastructure Roadmap discusses the country’s public digital infrastructure and data systems.
The 2026 Big Data Strategic Plan puts the Government Cloud alongside a Government Data Catalog and a National Big Data Platform. The catalog helps agencies find data and understand what is available; the platform is designed to bring information from multiple sources together for analysis. Public data can support shared services and research, while restricted data requires controlled access. Thailand’s data and AI readiness plans reflect the same push to reduce disconnected government systems.
For AI, shared infrastructure can lower the barriers to developing and running services across agencies. However, cloud capacity alone cannot make a model reliable. Agencies still need accurate, current data, clear ownership, and rules that specify who may access or reuse each dataset.
ThaiLLM and AI that understands Thai
ThaiLLM is a homegrown, open-weight Thai-language large language model initiative supported by the Digital Economy Development Fund. Its partner network includes the Big Data Institute (BDI), NECTEC, VISTEC, AIEAT, and AIAT. The initiative remains under development, so it should not be described as widely deployed or as outperforming other models.
Thai language use depends on context, including local expressions, spelling variations, and cultural references. A model trained and evaluated with suitable Thai data may better handle tasks such as summarizing Thai documents or answering questions about local services. That does not guarantee accuracy, though. The quality of its training data and the way teams test outputs matter just as much as language coverage.
Together, GDCC and national data platforms can supply shared foundations for public-sector AI, while ThaiLLM focuses on language-specific capability. Their value will depend on agencies pairing that infrastructure with dependable data and responsible access controls.
Where the government wants AI to make a difference
Thailand’s policy agenda is meant to move AI beyond isolated experiments and into public services and business applications that address practical needs. The 2026 Big Data Strategic Plan names six areas for wider data use, while the national AI strategy sets out the skills needed to put those applications into practice. These are policy priorities and possible use cases, not proof that projects are already operating.
Public services, health, tourism, and agriculture
The Big Data Strategic Plan identifies healthcare, tourism, environmental management, agriculture, trade, and economic development as priority sectors. In healthcare, agencies could use data to forecast demand for hospital services or help staff find relevant guidance in medical records. Any clinical use would need careful testing and human oversight; Thailand’s AI applications in healthcare offer further context.
Tourism data could help destinations estimate visitor flows and plan transport or staffing for busy periods. For environmental management, combining weather, land-use, and sensor data could support earlier warnings about floods or pollution. In agriculture, analysis of crop, soil, and weather information could help farmers and officials plan irrigation or identify emerging risks.
Trade and economic development also feature in the plan. For example, data analysis could help agencies spot changes in exports or evaluate which support programs reach businesses. The government’s Big Data Strategic Plan announcement sets out the six sectors and the goal of training at least 160,000 people in big data and AI skills by 2027.
Thai-language AI has another practical role: organizational knowledge retrieval. NSTDA, through NECTEC and in collaboration with the National Research Council of Thailand, is advancing Thai large language models for systems that help organizations find information in their own documents. That work could make internal policies and technical records easier to search, but accuracy and access controls remain essential.
AI skills for specialists and everyday workers
Thailand’s AI strategy describes three skill levels. Advanced researchers and developers build models and core technology; intermediate innovators and engineers adapt tools and create applications; basic workplace users apply AI services in their daily tasks. Each level fills a different need, from developing systems to using them responsibly.
The strategy’s earlier workforce framing included a target of 13,500 AI talents per year and a longer-term goal of more than 30,000 AI talents. Separately, the 2026 Big Data Strategic Plan aims to train at least 160,000 people in big data and AI skills by 2027. These targets cover different scopes: the larger figure includes broad workforce training, while the AI talent goals describe a narrower talent pipeline. They should not be added together or treated as equivalent.
How Thailand is balancing AI growth with safeguards
Thailand’s policy combines investment in AI capacity with a goal of responsible use. The national strategy and 2026 Big Data Strategic Plan set adopted priorities, while a proposed AI Act would add binding duties if lawmakers approve it.
The proposed AI Act and risk-based oversight
As of September 2026, Thailand’s AI Act was still a draft, not a law in force. ETDA released a revised draft on July 2 and held a public consultation through August 14. The proposal uses risk-based oversight, with stricter requirements for systems that pose greater risks. A summary of Thailand’s draft AI Act describes its proposed regulatory approach.
Reported provisions include limits on discriminatory profiling and covert manipulation, disclosure for AI-generated content and deepfakes, and impact assessments for certain systems. The draft also proposes a public registry of government AI systems. ETDA could be given authority to order corrective action, suspend services, or block systems, but those powers remain proposals until legislation takes effect.
The wider governance goal is to make AI use more accountable without treating every application as equally risky. Thailand’s AI governance guidelines for organizations offer context on current guidance, which is separate from the proposed Act.
For businesses, practical questions already matter: Can customers tell when AI generated or substantially changed content? Have teams checked whether a system treats groups unfairly? Does it handle personal data lawfully, and can a person review consequential decisions? Clear answers can help organizations prepare, even while the draft remains unsettled.
What progress means, and what remains uncertain
By 2026, Thailand had an adopted national AI strategy, a new big data plan, ThaiLLM development, and active discussions on AI legislation. These milestones show that policy work is moving on several fronts, but they don’t establish that public targets have been met or that AI services are producing reliable results.
The strategy material also doesn’t provide one consolidated figure for total AI spending. Delivery will depend on whether agencies can fund projects, maintain accurate and usable data, prepare workers, and set clear rules for oversight. For the public, the test is practical: when AI affects a service or decision, can people understand its role, question an error, and seek human review?
Thailand’s progress will be measured by the quality and accountability of systems put into use, not by plans or targets alone.
Frequently Asked Questions
Thailand’s AI plans combine an adopted policy framework with proposed regulation and active development projects. These answers clarify what is in place, what remains a goal, and what businesses should watch.
Is Thailand’s AI strategy a law?
No. The National AI Strategy and Action Plan 2022-2027 is a government policy plan approved by the Cabinet, not a binding AI statute. A separate AI Act was still under discussion as of September 2026, so businesses should distinguish its proposed requirements from laws already in force.
When does Thailand expect to meet its AI targets?
The current strategy sets 2027 as its planning horizon. That date marks the end of the plan’s target period, not a guarantee that every goal will be met by then. Progress depends on funding, agency capacity, workforce preparation, and the delivery of projects across sectors.
What is ThaiLLM?
ThaiLLM is Thailand’s national large language model initiative, designed to process and generate Thai text. Its focus on Thai-language capability supports applications that need to handle local language and context. The project also builds national AI capacity by developing domestic expertise and infrastructure.
How could Thailand’s AI policy affect businesses?
Businesses may gain access to shared infrastructure, a larger pool of trained workers, and support for wider AI adoption. At the same time, companies using AI must follow relevant privacy and security requirements, including rules for personal data. The proposed AI Act could add AI-specific duties if it becomes law; meanwhile, the OECD’s guidance for AI governance offers a reference for executives setting internal oversight practices.
Does Thailand have one central AI agency?
The National AI Committee coordinates the national direction, while ministries and technical agencies carry out different parts of the work. Ministries lead policy areas such as digital development, research, and education; agencies contribute technical expertise, infrastructure, and implementation. Thailand’s AI regulatory sandbox provides additional context on how agencies can test governance approaches with real-world projects.
