Taiwan AI Regulation Overview

Overview

Taiwan's approach to artificial intelligence regulation blends foundational legislative principles with national strategic initiatives, institutional evaluation capabilities, and municipal smart-city implementations. At the national level, the Artificial Intelligence Basic Act (promulgated and in force as of January 14, 2026) serves as the primary legal cornerstone, establishing a principles-first and government-first governance architecture. The framework emphasizes human-centric AI development, balancing technological innovation and economic growth against the protection of fundamental human rights, privacy, dignity, and cybersecurity.

Complementing this foundational law is the Taiwan AI Action Plan 2.0 (2023-2026), approved by the Executive Yuan in June 2023. The Action Plan allocates approximately NT$9-10 billion annually to leverage Taiwan's global semiconductor and ICT strengths, aiming to grow the domestic AI industry value past NT$250 billion by 2026. Furthermore, operational testing and trust verification are institutionalized through the AI Product and System Evaluation Centre (AIEC), established in December 2023 by the Ministry of Digital Affairs (MODA). Localized municipal implementations, such as the Taipei Smart City AI Initiative, showcase practical applications across municipal logistics, public health, disaster response, and citizen helplines.

Regulatory Approach

Taiwan adopts a dual-track, risk-based, and multi-layered regulatory approach. Rather than relying on a single rigid prescriptive statute for all commercial deployments, the Artificial Intelligence Basic Act outlines overarching fundamental principles while directing individual central competent authorities to establish risk-based management regulations and industry guidelines tailored to their respective sectors. This approach combines overarching national strategy with decentralized, sector-specific oversight, prioritizing government compliance before extending requirements to private enterprise.

The regulatory framework relies heavily on core ethical principles and technical evaluation systems. The Basic Act establishes seven core principles: sustainable development, human autonomy, privacy protection and data governance, cybersecurity and safety, transparency and explainability, fairness and non-discrimination, and accountability. To implement these principles, MODA is charged with formulating an internationally aligned AI risk classification framework, while the AIEC provides voluntary and phased verification based on ten key performance dimensions, ensuring that AI systems undergo systematic risk assessment and conformity testing.

Key AI Legislation

Taiwan's regulatory and policy framework for artificial intelligence comprises the following primary instruments recorded in the national database:

• Artificial Intelligence Basic Act (人工智慧基本法) (2026): Enacted on January 14, 2026, this foundational statute sets the legal framework, core principles, and institutional governance structure for human-centric AI development across Taiwan.
• Taiwan AI Action Plan 2.0 (2023-2026): Cabinet-level strategic blueprint approved by the Executive Yuan in June 2023, committing NT$9-10 billion annually to advance generative AI, semiconductor integration, talent cultivation, and research infrastructure.
• AI Product and System Evaluation Centre (AIEC) Framework (2023): Policy initiative established by the Ministry of Digital Affairs in December 2023, providing evaluation systems, technical testing, and certification for trustworthy AI products across ten performance categories.
• Taipei Smart City AI Initiative (2025): Municipal strategic framework launched by the Taipei City Government, guided by the 'Taipei City Government Guidelines for the Use of Artificial Intelligence' (2024) and the 'Three Arrows of AI' operational strategy.

Governance & Enforcement Bodies

Governance of AI in Taiwan is structured across cabinet-level coordination bodies, ministry-level regulators, specialized evaluation institutes, and local authorities. At the executive apex, the National Science and Technology Council (NSTC) serves as the central competent authority responsible for national AI development planning, research coordination, and drafting basic AI legislation. Overseeing whole-of-government implementation, the Executive Yuan mandates the establishment of a National AI Strategy Special Committee, chaired by the Premier, to review national AI programs and deliberate on strategic blueprints at least annually.

The Ministry of Digital Affairs (MODA) and its Administration for Digital Industries (ADI) hold primary responsibility for digital infrastructure, AI risk classification frameworks, and technical evaluation standards. MODA operates the AI Product and System Evaluation Centre (AIEC) in collaboration with the National Institute of Cyber Security and the Industrial Technology Research Institute (ITRI). Additionally, the Taiwan AI Center of Excellence (AICoE), opened in March 2024, aids in streamlining regulatory rule-making. At the municipal level, authorities such as the Taipei City Department of Information Technology (DOIT) oversee local AI adoption and compliance guidelines.

Penalties & Enforcement

The provided legal corpus contains no records detailing specific statutory monetary penalties, fine scales, or criminal sanctions for non-compliance under Taiwan's artificial intelligence laws. The Artificial Intelligence Basic Act establishes high-level mandates for internal control, public agency risk assessments, and sector-specific risk management regulations without listing explicit penalty schedules within the statutory summary.

Enforcement mechanisms in the current framework rely primarily on administrative governance, agency internal control norms, mandatory public sector risk assessments, and voluntary technical certification through the AIEC. Central competent authorities are instructed to issue sector-specific regulations and industry guidelines by January 2028, which may further define enforcement modalities within individual domains.

Data Protection Framework

Data protection and privacy governance within Taiwan's AI landscape are anchored by the Personal Data Protection Act and the core principles enshrined in the Artificial Intelligence Basic Act. The Basic Act emphasizes privacy protection and robust data governance, requiring strict adherence to the principle of data minimization, the safeguarding of personal data, and the protection of intellectual property rights while simultaneously fostering mechanisms for non-sensitive data sharing, open data, and data reuse to support AI model training.

The AI Product and System Evaluation Centre (AIEC) incorporates privacy protection into its ten-dimensional technical evaluation methodology. The AIEC specifically assesses AI systems for compliance with data protection requirements and privacy standards. Furthermore, the AIEC framework emphasizes linguistic diversity and data transparency by utilizing Taiwan's national languages and avoiding opaque data sources, enabling public involvement in determining training data inputs.

Sector-Specific Rules

Taiwan applies targeted rules and guidance across high-risk and public-facing sectors. Under the AIEC framework, high-risk AI applications are designated in sensitive domains such as healthcare, finance, and public safety, triggering enhanced technical scrutiny across safety, fairness, privacy, and explainability dimensions. Sectoral central competent authorities are required under the Artificial Intelligence Basic Act to formulate domain-specific risk management regulations and industry guidelines by January 2028.

In municipal public services and healthcare, specific operational rules have been established. Under the Taipei Smart City AI Initiative, municipal deployments are governed by the 'Taipei City Government Guidelines for the Use of Artificial Intelligence' (October 2024), regulating AI integration in citizen helplines (1999 Helpline), emergency response dispatch by the Fire Department, and smart traffic management. In healthcare, initiatives include pilot programs at Taipei City Hospital for brain cognitive health assessments, medical imaging diagnosis, and personalized treatment platforms.

International Alignment

Taiwan's AI regulatory and evaluation frameworks are designed to achieve close interoperability with international standards and Western regulatory models. The Ministry of Digital Affairs is explicitly mandated under the Basic Act to reference international benchmarks when creating Taiwan's AI risk classification framework, ensuring that domestic compliance systems can interface smoothly with global trade and technical regimes.

The AI Product and System Evaluation Centre (AIEC) aligns its testing protocols and evaluation guidelines with European Union AI Act requirements and United States AI governance frameworks. To promote global compatibility and trustworthy AI standards across the Asia-Pacific region, the AIEC collaborates with international technology firms, securing partnerships and support from global technology leaders including Meta and Microsoft.

Future Developments

Taiwan's AI legal and policy roadmap outlines several key statutory deadlines and strategic benchmarks. By March 31, 2026, the Ministry of Digital Affairs is scheduled to submit its AI Risk Classification Framework draft. Following this, the Executive Yuan is set to establish the National AI Strategy Special Committee by May 1, 2026, and all government agencies must complete public AI use risk assessments by July 1, 2026.

Further implementation benchmarks extend into subsequent years. Government agencies are mandated to complete internal control management norms for AI use by July 1, 2027. By January 1, 2028, sectoral central competent authorities must establish formal risk-based management regulations and industry guidelines for their respective industries. Strategically, Taiwan AI Action Plan 2.0 aims to achieve an AI industry market value exceeding NT$250 billion by the end of 2026.

Enforcement Bodies

Agency	Mandate	Key Powers	Website
National Science and Technology Council (NSTC)	Central competent authority for national AI policy, basic law drafting, and research coordination.	Formulates national AI blueprints, coordinates inter-ministerial resources, and oversees AI research.
Ministry of Digital Affairs (MODA) / Administration for Digital Industries (ADI)	Primary authority for digital infrastructure, AI risk classification frameworks, and evaluation systems.	Develops risk classification standards, oversees the AIEC, and provides technical evaluation tools and GPU resources.
Executive Yuan - National AI Strategy Special Committee	Cabinet-level strategic coordination body chaired by the Premier.	Deliberates and approves national AI guidelines, supervises inter-agency AI affairs, and conducts annual reviews.
AI Product and System Evaluation Centre (AIEC)	National evaluation body established by MODA, ITRI, and the National Institute of Cyber Security.	Evaluates and certifies AI products across ten technical dimensions, sets testing criteria, and conducts technical audits.
Department of Information Technology (DOIT), Taipei City Government	Executive municipal agency leading smart city AI policy and proof-of-concept programs in Taipei.	Implements municipal AI guidelines, manages PoC programs, and coordinates city-wide public AI deployment.