Central News Agency Report (Taipei, August 27, 2025)
The Executive Yuan is scheduled to discuss and approve the draft "Artificial Intelligence Basic Act" tomorrow. In addition to clearly defining the basic principles for the research and application of AI, the draft also stipulates that AI applications should be prevented from causing illegal activity, unnecessary collection of personal data should be avoided, and unemployed workers displaced by AI should receive vocational assistance. Furthermore, the Ministry of Digital Affairs (MODA) is mandated to promote an internationally aligned AI risk classification framework that references international standards or regulations.


The highly anticipated draft of the "Artificial Intelligence Basic Act" was initially proposed by the National Science and Technology Council (NSTC). However, due to subsequent considerations regarding the rights and interests of the public, including national security and the management of high-risk AI, the draft was reorganized by the Ministry of Digital Affairs (MODA) before being submitted for review by the Executive Yuan. It will be discussed and potentially approved by the Executive Yuan Council tomorrow.

Core Principles and Development Mandates

The draft clearly mandates that the research and application of AI must prioritize social welfare and digital equity while developing sound governance. It must adhere to principles including: sustainability, human autonomy, privacy protection and data governance, safety, transparency and explainability, and accountability.
To actively promote AI research, application, and infrastructure development, the draft stipulates that the government shall properly plan the overall allocation of resources. This includes providing subsidies, commissions, investments, incentives, and assistance for AI-related industries, as well as offering fiscal incentives such as tax and financial breaks. Relevant competent authorities are also permitted to establish or enhance innovative experimental environments for AI research and application services tailored to cutting-edge AI products or services, thereby facilitating technological innovation and sustainable development in AI.

Preventing Negative Impacts and High-Risk Management

In addition to actively promoting AI development, the draft aims to mitigate the negative consequences of AI applications. It mandates that the government should prevent AI applications from causing harm to people's lives, bodies, liberties, properties, social orders, national security, or the ecological environment. It also aims to address issues such as conflict of interest, bias, discrimination, false advertising, misleading or fabricated information, and violations of relevant laws, including the Child and Youth Welfare and Rights Protection Act, Fair Trade Act, Consumer Protection Act, and Personal Data Protection Act. The government may use legislation to limit or prohibit such occurrences.
The draft stipulates that the Ministry of Digital Affairs shall promote an internationally aligned Artificial Intelligence Risk Classification Framework, referencing international standards or regulations. Competent authorities should utilise this framework to formulate risk-based hierarchical management regulations, based on the need for AI application risk management.

Workforce Balance and Data Governance

To address the balance between AI and the workforce, the draft mandates that the government shall provide vocational assistance to those unemployed due to the utilization of AI, based on their work capabilities.
The draft requires the government to clarify liability and attribution conditions for high-risk AI applications, and to establish related remedy, compensation, or insurance mechanisms. Furthermore, the personal data protection authority must assist competent authorities in avoiding unnecessary collection, processing, or utilization of personal data during AI research and application. It should also promote measures and mechanisms that incorporate privacy by design to protect the rights and interests of data subjects.
Additionally, the draft stipulates that the government should establish mechanisms for data openness, sharing, and reuse to enhance the availability of data for AI applications. It must regularly review and adjust relevant laws and regulations. Concurrently, efforts should be made to enhance the quality and quantity of data used for AI in Taiwan, ensuring that AI training and output results align with the nation's diverse cultural values and protect intellectual property rights. (Editor: Lin Hsing-meng) 1140827
 
Back to the top