Liberty Times Net, December 23, 2025
By Hsieh Chun-lin / Taipei
The Legislative Yuan today passed the “Artificial Intelligence Basic Act” on its third reading. In addition to clearly defining “artificial intelligence” (AI), the Act mandates that the Executive Yuan establish a “National Artificial Intelligence Strategy Special Committee” to coordinate, promote, and supervise AI-related affairs nationwide. However, despite repeated statements by the Democratic Progressive Party (DPP) caucus that basic acts have never set a competent authority by legislative precedent, the Kuomintang (KMT) and the Taiwan People’s Party (TPP) jointly passed an amendment stipulating that the National Science and Technology Council (NSTC) shall serve as the central competent authority, while the competent authorities at the local level shall be the special municipalities and county/city governments.
The “Artificial Intelligence Basic Act” underwent three joint reviews by the Legislative Yuan’s Education and Culture Committee and Transportation Committee, completed its preliminary review on August 4, followed by consultations on August 21 and further inter-party negotiations convened by President of the Legislative Yuan, Han Kuo-yu, on December 4. The bill completed the third-reading process today.
Regarding the definition of “artificial intelligence”, the provisions adopted on third reading define it as a system with autonomous operational capability that, through input or sensing and by means of machine learning and algorithms, generates outputs—such as predictions, content, recommendations, or decisions—aimed at explicit or implicit objectives and capable of affecting physical or virtual environments.
The new Act also stipulates that, in promoting AI research, development, and application, the government shall, on the premise of safeguarding the public interest, digital equity, innovation, and national competitiveness, adhere to seven core principles: sustainability and well-being, human autonomy, privacy protection and data governance, cybersecurity and safety, transparency and explainability, fairness and non-discrimination, and accountability.
To prevent harm arising from AI applications, the provisions adopted on third reading require the government to avoid AI uses that infringe upon people’s life, bodily integrity, freedom, or property; undermine social order, national security, or the ecological environment; or involve bias, discrimination, false advertising, misinformation, or fabrication in violation of relevant laws and regulations.
In addition, the government shall act in accordance with the best interests of children and youth. AI products or systems identified as high-risk applications must clearly display precautionary notices or warnings. The government shall also provide or recommend assessment and verification tools or methods, the development of which shall involve consultation with relevant stakeholders, industry and academia, civil society organizations, and legal experts.
The provisions adopted on third reading further provide that the Executive Yuan shall establish the National Artificial Intelligence Strategy Special Committee, convened by the Premier of the Executive Yuan and composed of scholars and experts, industry representatives, heads of relevant agencies, and mayors or county/city heads, to formulate national AI development guidelines. The committee shall convene at least once annually, with secretariat operations handled by the NSTC.
With respect to budgeting, the new Act specifies that, within the limits of fiscal capacity, the government shall allocate sufficient budgets, adopt necessary measures, and actively promote AI research, application, and infrastructure development.
Regarding personal data protection, the provisions adopted on third reading require the competent authorities of various sectors to consult with the personal data protection authority during AI research and application processes, in order to avoid unnecessary collection, processing, or use of personal data. They shall also promote the incorporation of privacy protection into default settings and system design, so as to safeguard the rights and interests of data subjects.
After the bill passed its third reading, KMT Legislator, Ko Ju-chun, stated that the passage of the Act marks only a beginning rather than an endpoint, and that it will still be necessary to accelerate the development of supporting legislation to address long-standing institutional bottlenecks faced by industry. DPP Legislator, Chang Ya-lin, remarked that this is not merely the enactment of a law, but also an important milestone marking Taiwan’s formal step forward in confronting the wave of technological advancement.