We sincerely thank all the distinguished guests for participating in the 7th International Conference on Artificial Intelligence and Law.

"The rapid transformation of AI technology profoundly impacts industries, government policies, and people's lives and behaviors, and AI law is intimately connected to the future of life." In his address at the conference, Mr. Chu Chao-Min, Chairman of the International Research Foundation for Artificial Intelligence Law, emphasized that over the seven years since its establishment, the Foundation has been dedicated to advancing AI law. The draft AI Basic Act prepared by the Foundation was submitted to the Legislative Yuan for review this March, with the hope that it will serve as the backing for the evolution of AI law in Taiwan.


翻譯如下:

Dr. Iris Chang, Director of the Stanford Taiwan Science and Technology Center, also stated that AI has been an important topic since the center's establishment in 2023. Stanford University plays a crucial pivotal role in Silicon Valley, and besides focusing on technology, another important mission is to guide the formulation of AI policy and further expects to maximize the benefits AI brings to humanity through the promotion of legislation.

Dr. Chen-Kang Su, Deputy Minister of the National Science and Technology Council (NSTC), pointed out that the government currently has two major directions for promoting AI development: first, strengthening infrastructure such as computing power and network communication environment; and second, legal framework construction. These two directions complement each other to jointly promote the development of the AI industry, enabling all sectors to connect with AI. At the level of legal framework construction, the formulation of the AI Basic Act is extremely important.

翻譯如下:

Li-Ching Chang, CEO of the International Research Foundation for Artificial Intelligence Law and Advisor to the Executive Yuan's Digital Policy Legal Platform, presented a comparative analysis of guidelines and regulations from the European Union, the United States, and the OECD. She also offered recommendations regarding the shortcomings of Taiwan's current draft Artificial Intelligence Basic Act. Chang analyzed that the EU adopts a relatively strict regulatory approach, while the US relies on voluntary guidelines, encouraging corporate participation and regulatory adaptation, with a particular focus on AI within the national security system. Taiwan's current two versions of the draft Basic Act both adopt a risk-based hierarchical supervision approach. However, the non-governmental version places greater emphasis on industry participation, highlighting industry self-regulation and international alignment, and integrates the national policy of "building a smart nation." She stated that the content of the Basic Act must balance the dual objectives of promoting AI development and governance. She suggested a fusion of the EU and US models: implementing risk-based governance with the binding force of law while retaining policy flexibility and establishing a comprehensive oversight mechanism to achieve a balance between technology and innovation, thus fostering the diversified development of the AI industry.

翻譯如下:

Us Enhanced U.S. Export Control Enforcement: How Taiwan's AI Supply Chain Should Respond
In addition to discussions on the AI Basic Act, Chen-Tso Chu, Secretary General of the International Research Foundation for Artificial Intelligence Law, Vice Chairman of the Taiwan Bar Association, and currently a Visiting Scholar at Stanford University, shared insights on the risks posed by U.S. export control regulations to Taiwan's industry and strategies for response, from the perspective of Taiwan's AI industry development.

As the U.S. increasingly strengthens its export control regulations, particularly the blockade targeting products and technology involving China's AI technology and semiconductor industry, the role of Taiwanese companies in the global AI supply chain has become more sensitive and complex. Chu suggested that companies should establish an internal review process referencing the U.S. BIS (Bureau of Industry and Security) codes. This process should first confirm the controlled technology items, second confirm the reasons for control, and then confirm the controlled regions and recipients, including transshipment agents and end-users. Finally, they must also confirm the control pathways, as the global trade supply chain may redirect chips to controlled entities through re-export or even domestic transfer. Furthermore, although the BIS has exemption clauses, any high-computing power AI semiconductor chip destined for controlled countries or controlled entities cannot be exempted.

In response to U.S. export control regulations, Chu also offered three recommendations for Taiwanese companies when formulating future development strategies:

Prioritize the relocation of high-risk production lines and engineering and R&D teams that handle core technology when moving the industrial chain.

Establish a liaison window based on BIS to anticipate future BIS policy direction.

Chu also cautioned that, as the industrial chain follows the "China + 1" trend, selecting destination countries for supply chain relocation that offer preferential tariffs to reduce costs will be a crucial strategy for dealing with export controls.