Shifting AI Governance in East Asia: AI Legislative Progress in Japan, South Korea and Taiwan
Shifting AI Governance in East Asia: AI Legislative Progress in Japan, South Korea and Taiwan
2025/09/09
Keywords: artificial intelligence, artificial intelligence regulation
I.Introduction
The landscape of AI governance in East Asia is changing, with two new AI laws enacted and one on the way. In South Korea, an act titled “the Basic Act on the Development of Artificial Intelligence and the Establishment of Foundation for Trustworthiness“ (“인공지능 발전과 신뢰 기반 조성 등에 관한 기본법”, henceforth referred to as “South Korea’s AI Act” or “SKAIA”)[1]was approved on December 26[2], 2024 and promulgated on January 21, 2025. The AI Basic Act is designed to establish a national AI governance framework and systematically foster the AI industry while preventing potential AI risks.[3] A few months later, Japan’s first law regulating AI was passed by the National Diet on May 28, 2025. The new law is titled "the Act on Promotion of Research and Development, and Utilization of AI-related Technology" (“人工知能関連技術の研究開発及び活用の推進に関する法律”, henceforth referred to as "Japan's AI Act" or "JAIA")[4], which reflects the strong will of the government to catch up in the global AI race.[5] Elsewhere in the region, Taiwan’s Executive Yuan finally passed its draft AI Basic Act (“人工智慧基本法草案”) on August 28[6] [7], which must now be submitted to the Legislative Yuan for deliberation. The government hopes the new law will lay the foundation for establishing Taiwan as an AI island and a key driving force in global AI development.[8] This article will give a quick overview of the key features of the three new AI regulations to illustrate the new landscape these countries are shaping in AI governance.
II.Key features of Japan’s AI Act (JAIA)
1.Purpose and principles of JAIA
Given Japan's lagging AI development and rising public concerns, JAIA reflects the government's worry about falling behind global peers in AI investment and adoption.[9] It is believed that new laws are needed in addition to existing laws and regulations to promote innovation and address risks.[10] Hence JAIA aims to advance the R&D and application of AI through the formulation of basic principles and plans, and the establishment of an "AI Strategic Headquarters".[11]
JAIA establishes basic principles for the promotion of the R&D and application of AI-related technologies[12], including enhancing industry R&D capabilities and competitiveness, systematically promoting AI collaboration from research to application with transparency, and enabling Japan to shape global norms through international cooperation.[13]
2.Industry Development and Promotion
JAIA requires the government to develop a National AI Basic Plan, in accordance with the basic principles, to promote the R&D and application of AI. The AI Basic Plan should set out fundamental policy guidelines and measures to comprehensively and systemically advance the R&D and application of AI-related technologies, along with other necessary provisions.[14]
JAIA also specifies basic measures to be included in the plan, which cover issues of promotion of R&D, expansion and sharing of facilities and data, human resources and education, international engagement in AI norm setting, and domestic guidelines making. In addition, the government should monitor AI technology trends and analyze cases of rights violations from improper AI use to develop countermeasures and provide guidance accordingly.[15]
3.Governance
JAIA stipulates that an AI Strategy Headquarters should be established under the Cabinet, composed of all cabinet members and headed by the Prime Minister.[16] The AI Strategic Headquarters is tasked with comprehensively and systematically advancing AI-related technology R&D and application policies, including the formulation, promotion, and implementation of AI Basic Plans and other related initiatives.[17] The Act also empowers the AI Strategy Headquarters to invite stakeholders to provide information, opinions or explanations, and other necessary assistance.[18]
4.Risk managements and rights protection
JAIA does not impose direct compliance obligations, but AI companies and research institutions are required to cooperate with government investigations and follow government guidance in cases involving violations of human rights and interests.[19]
5.Implementation of JAIA and Follow-up Work
JAIA came into force in May 2025. The Japanese government is required to develop guidelines that align with international standards and launch the Strategic Headquarters for the preparation and implementation of the National AI Basic Plan.
III.Key features of the South Korea’s AI Act (SKAIA)
1.Purpose and principles of SKAIA
SKAIA is designed to establish a foundation for AI development and trustworthiness, increasing citizens’ rights and interests protection, quality of life, and the country’s competitiveness.[20] It focuses on advancing national AI collaboration to foster a flourishing AI sector and developing legal frameworks to mitigate risks.[21]
Accordingly, the Act establishes basic AI development principles: prioritizing safety and reliability to improve quality of life, and ensuring those affected by AI output receive clear, meaningful explanations within reasonable parameters.[22]
2.Industry development and promotion
Supporting AI technology and industry development is a key feature of SKAIA. It establishes comprehensive measures covering technology development, industry revitalization, SME support, industrial foundations, talent cultivation, regulatory adaptation, and international cooperation.[23]
3.Governance
SKAIA also strengthens the institutional framework for AI governance. The Ministry of Science and ICT (henceforth referred to as “MSIT”) is mandated to execute an AI Master Plan every three years and empowered to investigate violations, require corrective action, and impose fines on non-compliant entities.[24]
The National AI Committee is authorized to review and decide on the AI Master Plan and AI-related matters, making it the highest decision-making body for South Korea's AI policies. It is composed of the heads of central administrative agencies and civilian AI experts appointed by the president.[25]
SKAIA also establishes the AI Policy Center to support MSIT on AI policy formulation, and the AI Safety Institute for AI safety matters.[26]
4.Risk management and rights protection
SKAIA imposes specific obligations on operators of high-impact AI and generative AI systems. All operators must ensure system transparency and safety, while high-impact AI operators face additional responsibilities including conducting fundamental rights impact assessments.[27]
High-impact AI systems are defined as AI systems that have a significant impact on or may pose a risk to human life, safety, and fundamental rights and are mainly utilized in critical infrastructure sectors and human rights-sensitive areas, or other areas specified by presidential decree.[28] The procedure for determining whether an AI system qualifies as high-impact AI will be established through subordinate legislation.[29]
5.Implementation of SKAIA and Follow-up Work
SKAIA will come into effect on January 1, 2026 and the formulation of subordinate statutes that detail enforcement mechanisms and guidelines should be expedited. However, domestic critics argue that corporate obligation provisions may hinder AI development and advocate for postponing their implementation.[30] Actually, an amendment to the Act was proposed in April 2025, seeking such a postponement along with a three-year grace period.[31]
IV. Key features of Taiwan’s draft AI Basic Act
1.Purpose and Principles of the draft AI Basic Act[32]
Taiwan adopts a relatively conservative approach to AI policy and measures to boost industrial development have long occupied the agenda of AI governance. Given that AI is a crucial technology for national development, the draft AI Basic Act (henceforth referred to as "the draft Act") seeks to ensure that AI technology develops vigorously in a human-centered approach, encourage innovation while considering human rights, and safeguard Taiwan’s national sovereignty and cultural values.[33]
Hence, the draft Act establishes seven guiding principles in line with international norms, which are sustainability, human autonomy, privacy protection and data governance, security, transparency and explainability, fairness and accountability.[34]
2.Industry Development and Promotion
It is the government’s responsibility to promote the R&D and application of AI and construct the infrastructure needed.[35] In order to facilitate AI innovations, competent authorities may provide a controlled environment for testing and validating AI innovation products and services before they are released to the market or put into use.[36] Considering the wide scope of AI application and development, the government is encouraged to collaborate with the private sector, including through public-private partnerships, and should promote international cooperation on AI matters.[37] The government should also continue to comprehensively promote AI education at all levels to enhance the public's AI literacy.[38]
Data is crucial for AI development, so the draft Act mandates the government to establish mechanisms to enhance data availability, and measures to facilitate AI outputs that maintain the country's multicultural values, and protect intellectual property rights.[39]
3.Risk Management and Rights Protection
(1) Risk Management
The draft Act includes several provisions addressing AI risks. The government should take steps to prevent AI from being used for illegal purposes. For example, Ministry of Digital Affairs (MODA) and other relevant agencies may provide or recommend tools or methods for AI evaluation and verification to avoid misuse of AI.[40] Secondly, MODA is mandated to foster an AI risk classification framework, based on which sectoral competent authorities should establish risk-based tiered management standards.[41] Thirdly, the government may, through binding regulations or non-binding administrative guidance, promote safety standards, verification, transparent and explainable traceability, or accountability mechanisms to enhance the trustworthiness of AI development and application.[42] Lastly, the government should clarify the ownership and conditions of liability for high-risk AI applications and establish relevant mechanisms for relief, compensation or insurance to protect affected parties.[43] However, AI application responsibility norms would not apply to pre-release activities in order to support technological innovation.[44] [45]
(2) Rights Protection
The draft Act concerns not only the privacy rights of individuals but also labor rights. The government should ensure the protection of personal data used throughout the AI lifecycle on the one hand[46] , and also protect workers' rights and provide necessary assistance to help them adapt to technological changes, especially those who have lost their jobs due to AI use.[47]
4.Governance and Implementation
Despite the heated debate regarding the designation of a dedicated AI regulatory authority in the country, the Executive Yuan decided against establishing such an authority, given AI's cross-ministerial nature. Relevant competent authorities will be responsible for formulating implementing regulations and guidelines and the Executive Yuan will continue to guide relevant agencies and departments at all levels through the existing Digital Legal Coordination Meeting to facilitate the development of AI.[48]
V.Analysis and conclusion
Japan, South Korea and Taiwan all seek to maintain the countries' momentum in promoting AI development through AI legislation. The three parties all emphasize trustworthy AI, though they actually place greater emphasis on AI development. They share considerable common ground in the policies to foster AI industry development, such as promoting AI R&D and application and supporting infrastructure-building, and diverge in their approaches to addressing potential AI-related risks and governance structure.
Japan adopts a ‘light touch’ regulatory approach to AI regulation, maintaining coherent policy coordination that responds to domestic imperatives and global trends without imposing regulatory burdens on industries.[49] The country favors a soft approach with governmental guidance. In contrast, South Korea incorporates regulatory provisions specifically targeting high-impact AI systems in its AI Basic Act, seeking to balance between enhancing national competitiveness through AI and mitigating potential risks stemming from AI misuse, though this approach actually faces some domestic opposition currently. Taiwan adopts an approach similar to Japan's. The draft AI Basic Act avoids imposing regulatory obligations, and the government will prioritize AI verification and evaluation mechanisms to ensure trustworthy AI development.
Regarding governance approaches, both Japan and South Korea seek to strengthen governmental AI governance functions through legislation, with Japan establishing an AI Strategic Headquarters and South Korea creating an AI Committee, both operating under their respective Cabinets. In contrast, Taiwan's draft AI Basic Act does not address governance structural matters.
Given the profound societal transformations that AI technology may bring, all three East Asian countries recognize the importance of sustained AI advancement while acknowledging the critical need to ensure AI safety and trustworthiness to protect human rights. In an era of intense global AI competition, it seems to be the best policy for governments to carefully design AI policies that strike a balance between fostering innovation and safeguarding human rights. This cautious approach is essential as significant challenges remain and AI risks demand comprehensive solutions.
Reference:
[1] 인공지능 발전과 신뢰 기반 조성 등에 관한 기본법（법률 제20676호, 2025. 1. 21, 제정），법제처 국가법령정보센터，https://www.law.go.kr/%EB%B2%95%EB%A0%B9/%EC%9D%B8%EA%B3%B5%EC%A7%80%EB%8A%A5%20%EB%B0%
9C%EC%A0%84%EA%B3%BC%20%EC%8B%A0%EB%A2%B0%20%EA%B8%B0%EB%B0%98%20%EC%A1%B0%EC%
84%B1%20%EB%93%B1%EC%97%90%20%EA%B4%80%ED%95%9C%20%EA%B8%B0%EB%B3%B8%EB%B2%95/(206
76,20250121) （最後瀏覽日：2025/09/11）。
[2] A New Chapter in the Age of AI: Basic Act on AI Passed at the National Assembly‘s Plenary Session, Ministry of Science and ICT, https://www.msit.go.kr/eng/bbs/view.do?sCode=eng&mId=4&mPid=2&pageIndex=&bbsSeqNo=42&nttSeqNo=1071&searchOpt=ALL&searchTxt= (last visited Sept. 11, 2025).
[3] A New Chapter in the Age of AI: Basic Act on AI Passed at the National Assembly‘s Plenary Session, Ministry of Science and ICT, https://www.msit.go.kr/eng/bbs/view.do?sCode=eng&mId=4&mPid=2&pageIndex=&bbsSeqNo=42&nttSeqNo=1071&searchOpt=ALL&searchTxt= (last visited Sept. 11, 2025).
[4] 人工知能関連技術の研究開発及び活用の推進に関する法律（令和7年法律第53号），e-Gov法令検索，https://laws.e-gov.go.jp/law/507AC0000000053（最後瀏覽日：2025/09/11）。
[5] CABINET OFFICE, GOVERNMENT OF JAPAN, Outline of the Act on Promotion of Research and Development, and Utilization of AI-related Technology (AI Act), https://www8.cao.go.jp/cstp/ai/ai_hou_gaiyou_en.pdf (last visited Sept. 11, 2025).
[6] 〈政院通過「人工智慧基本法」草案 建構AI發展與應用良善環境 打造臺灣成為AI人工智慧島〉，行政院，https://www.ey.gov.tw/Page/9277F759E41CCD91/5d673d1e-f418-47dc-ab35-a06600f77f07（最後瀏覽日：2025/09/09）。
[7] There are other AI bills brought up by legislators in the Legislative Yuan. The purpose of this article is to analyze the AI governance priorities of the governments of Japan, South Korea, and Taiwan; therefore, other AI bills proposed by legislators are not included in the discussion.
[8] 蘇文彬，〈行政院通過AI基本法草案，將不設立AI專責機關〉，iThome，2025/08/28，https://www.ithome.com.tw/news/170874 (最後瀏覽日：2025/09/09)。
[9] Japan’s AI Bill Advances Toward Enactment, Connect on Tech (May 27, 2025), https://connectontech.bakermckenzie.com/japans-ai-bill-advances-toward-enactment/ (last visited Sept. 9, 2025).
[10] 松尾剛行，〈【2025年施行】AI新法とは？AIの研究開発・利活用を推進する法律を分かりやすく解説！〉，Keiyaku-Watch，https://keiyaku-watch.jp/media/hourei/2025-ai-law/（最後瀏覽日：2025/09/11）。
[11] 人工知能関連技術の研究開発及び活用の推進に関する法律（令和7年法律第53号）第1条。
[12] 人工知能関連技術の研究開発及び活用の推進に関する法律（令和7年法律第53号）第3条。
[13] Japan Enacts AI Promotion Act: Overview and Implications for Businesses, Zelo Law Square (May, 2025), https://zelojapan.com/en/lawsquare/56899 (last visited Sept. 9, 2025).
[14] 人工知能関連技術の研究開発及び活用の推進に関する法律（令和7年法律第53号）第18条。
[15] 人工知能関連技術の研究開発及び活用の推進に関する法律（令和7年法律第53号）第11-17条。
[16] 人工知能関連技術の研究開発及び活用の推進に関する法律（令和7年法律第53号）第19、21-24条。
[17] 人工知能関連技術の研究開発及び活用の推進に関する法律（令和7年法律第53号）第20条。
[18] 人工知能関連技術の研究開発及び活用の推進に関する法律（令和7年法律第53号）第25条。
[19] 人工知能関連技術の研究開発及び活用の推進に関する法律（令和7年法律第53号）第16条。
[20] 인공지능 발전과 신뢰 기반 조성 등에 관한 기본법，제1조。
[21] The Korean AI Basic Act: Asia’s First Comprehensive Framework on AI, Lexology (Mar. 17, 2025), https://www.lexology.com/library/detail.aspx?g=f91ff0fb-94ed-4aa9-b667-65d6206a7227 (last visited Sept. 9, 2025).
[22] 인공지능 발전과 신뢰 기반 조성 등에 관한 기본법，제3조。
[23] 인공지능 발전과 신뢰 기반 조성 등에 관한 기본법，제13-26조。
[24] 인공지능 발전과 신뢰 기반 조성 등에 관한 기본법，제40조。
[25] 인공지능 발전과 신뢰 기반 조성 등에 관한 기본법，제7조。
[26] 인공지능 발전과 신뢰 기반 조성 등에 관한 기본법，제6-12조。
[27] 인공지능 발전과 신뢰 기반 조성 등에 관한 기본법，제31-32조。
[28] 인공지능 발전과 신뢰 기반 조성 등에 관한 기본법，제4조。
[29] 인공지능 발전과 신뢰 기반 조성 등에 관한 기본법，제33조。
[30] Seungmin (Helen) Lee, South Korea’s Evolving AI Regulations, Stimson (June 12, 2025), https://www.stimson.org/2025/south-koreas-evolving-ai-regulations/ (last visited Sept. 9, 2025).
[31] 〈인공지능 발전과 신뢰 기반 조성 등에 관한 기본법 일부개정법률안〉，대한민국국회，https://likms.assembly.go.kr/bill/bi/billDetailPage.do?billId=PRC_N2M5K0S3R2R0Q1O3X5X1W1U1T7P3Q6&currMenuNo=2600044（最後瀏覽日：2025/09/09）。
[32] 〈政院通過「人工智慧基本法」草案 建構AI發展與應用良善環境 打造臺灣成為AI人工智慧島〉，行政院，https://www.ey.gov.tw/Page/9277F759E41CCD91/5d673d1e-f418-47dc-ab35-a06600f77f07（最後瀏覽日：2025/09/09）。
[33] 人工智慧基本法草案第1條。
[34] 人工智慧基本法草案第3條。
[35] 人工智慧基本法草案第4條。
[36] 人工智慧基本法草案第5條。
[37] 人工智慧基本法草案第6條。
[38] 人工智慧基本法草案第7條。
[39] 人工智慧基本法草案第14條。
[40] 人工智慧基本法草案第8條。
[41] 人工智慧基本法草案第9條。
[42] 人工智慧基本法草案第10條。
[43] 人工智慧基本法草案第11條。
[44] 人工智慧基本法草案第11條。
[45] See also: Taiwan Rolls Out Draft Artificial Intelligence Law, OCACNEWS, July 18, 2024, https://ocacnews.net/article/374412 (last visited Sept. 3, 2025).
[46] 人工智慧基本法草案第14條。
[47] 人工智慧基本法草案第12條。
[48] 蘇文彬，〈行政院通過AI基本法草案，將不設立AI專責機關〉，iThome，2025/08/28，https://www.ithome.com.tw/news/170874 (最後瀏覽日：2025/09/09)。
[49] Sun Ryung Park, Less Regulation, More Innovation in Japan’s AI Governance, East Asia Forum (May 21, 2025), https://eastasiaforum.org/2025/05/21/less-regulation-more-innovation-in-japans-ai-governance/ (last visited July 4, 2025).
A Preferred Model for Taiwan’s agency level AI risk categorization and management: A Cross-Jurisdictional Perspective 2025/09/15 Taiwan’s draft Artificial Intelligence Basic Law includes a provision allowing each government agency to establish its own risk-based AI management rules tailored to sector-specific regulatory needs[1]. To strike an effective balance between innovation and oversight, selecting an appropriate reference model is essential. After comparing major jurisdictions, this research argues that the United States Office of Management and Budget (OMB) Memorandum M-25-21—Accelerating Federal Use of AI through Innovation, Governance, and Public Trust[2]—offers the most balanced and practical approach for Taiwan’s agencies to refer to at this initial stage of developing AI regulation and promoting AI adoption. This article will first present an overview of the U.S. M-25-21 framework and its key features. It will then explain why the U.S. model is more suitable for Taiwan than those of other jurisdictions. Finally, it will conclude with recommendations for the government. I. Overview of the U.S. M-25-21 Framework Issued in April 2025 under Executive Order 14179, M-25-21 directs federal agencies to accelerate the adoption of artificial intelligence while maintaining a set of minimum safeguards. The memorandum identifies three priorities—innovation, governance, and public trust—and structures AI oversight around these principles. It requires every executive branch agency to designate a Chief AI Officer (CAIO), a senior official empowered to promote AI innovation, maintain a current inventory of AI use cases, and ensure that processes such as determining “high-impact” uses are in place. Rather than imposing a centralized management system, M-25-21 allows each agency to make context-sensitive determinations and to accept or waive risk management requirements. This approach recognizes that agencies vary widely in mission and capacity and are best positioned to understand the potential risks and benefits of AI within their own domains. The memorandum defines high-impact AI as systems whose outputs serve as a principal basis for decisions or actions with legal, material, binding, or significant rights and safety consequences. It offers a non-exhaustive list of presumed high-impact categories, including safety-critical functions of critical infrastructure, traffic management, patient diagnosis, blocking protected speech, and law enforcement applications. If an agency official determines that a specific AI use within these categories does not meet the high-impact definition, they must submit written documentation to notify the CAIO. By tying the definition to the effect of an AI system’s output rather than to a fixed sectoral list, M-25-21 provides a flexible method for identifying high-risk AI applications while preserving room for innovation. II. Key Features of the U.S. M-25-21 Framework A. Minimum Risk Management Practices To ensure protection without creating excessive barriers, M-25-21 specifies a set of minimum risk management practices that each agency must apply when using high-impact AI. Agencies are required to conduct pre-deployment testing under realistic conditions to confirm that AI systems perform as intended and to prepare appropriate risk mitigation plans. Even when agencies lack access to source code or training data, they are expected to use alternative testing methods—such as querying the AI service and observing its outputs—to assess performance and potential risks. Before deploying a high-impact AI system, agencies must complete an AI impact assessment. This assessment must explain the system’s intended purpose and expected benefits, analyze the quality and appropriateness of the data used, and evaluate potential impacts on privacy, civil rights, and civil liberties. It should also include a cost analysis, planned reassessment schedules and procedures, and comments highlighting potential concerns or gaps from an independent reviewer who was not involved in the system’s development. Importantly, the assessment must carry the signature of an accountable official who formally accepts the risk of deploying the AI system. Once deployed, agencies are expected to monitor AI systems continuously for performance drift, security vulnerabilities, or unforeseen adverse effects, and to implement appropriate mitigations and maintain documentation. Human oversight is equally essential: operators must receive specific training to interpret AI outputs, intervene when necessary, and use fail-safes or override mechanisms to minimize the risk of significant harm in high-impact situations. To protect the public, M-25-21 insists that individuals affected by AI-enabled decisions have access to timely human review and opportunities to appeal adverse outcomes. Appeals should not impose unnecessary burdens on individuals or the administration. Furthermore, agencies are expected to seek feedback from end users and the public to inform AI-related decision-making. These combined practices—testing, assessment, independent review, monitoring, human oversight, remedies, and feedback—form a balanced foundation for responsible AI use. The memorandum also requires agencies to safely discontinue any high-impact use cases that fail to comply with the minimum practices. B. Waiver System: Purpose and Conditions A distinctive feature of M-25-21 is its formal system for waivers from the minimum risk management practices. The waiver mechanism exists to reconcile two priorities: ensuring safety and rights protections on the one hand, and enabling innovation and rapid response on the other. Waivers may be considered when following a particular requirement would actually increase risks to safety or rights overall, or when compliance would create an unacceptable impediment to critical agency operations. For example, during a natural disaster or public health emergency, strict adherence to every procedural requirement might delay the deployment of an AI application that could save lives. In such situations, the CAIO may authorize a waiver to permit rapid deployment while still tracking and reassessing the use. Waivers for pilot programs are equally important for encouraging experimentation and innovation. They allow agencies to conduct small-scale, time-limited AI projects without implementing all minimum risk management practices, provided certain conditions are met: the pilot must be certified by the CAIO, centrally tracked, offer opt-in and opt-out options for individual participation, and apply minimum risk management practices where practicable. The memorandum imposes safeguards on this flexibility. Every waiver must be documented with a written determination explaining the reasoning, centrally tracked, and reassessed annually or whenever significant changes to the AI application’s conditions or context occur. CAIOs retain the power to revoke waivers at any time, and agencies must report any granted or revoked waiver to OMB annually and within 30 days of significant modifications. This approach maintains accountability while preventing rigid rules from becoming obstacles to effective governance. C. Disclosure Requirements for High-Impact Use and Waivers M-25-21 strongly emphasizes transparency as a pillar of public trust. Each agency must maintain an inventory of all AI use cases, submit it to OMB, and post a public version on the agency’s website. This inventory should be updated annually and, ideally, throughout the year to reflect the agency’s current use of AI. Transparency ensures that the public, civil society, and oversight bodies can understand where AI is influencing important government decisions without exposing sensitive or classified details. Similarly, agencies must publicly release summaries of each waiver or determination, including the justification, or explicitly indicate when no determinations or waivers are active. By making these summaries visible, the system builds confidence that waivers are granted for legitimate reasons. At the same time, OMB retains the authority to request detailed records concerning exception determinations within presumed high-impact categories. This combination of public disclosure and federal oversight helps maintain trust while safeguarding privacy, national security, and proprietary information. III. Why M-25-21 Stands Out for Taiwan’s AI Governance among Global Approaches Taiwan’s draft AI Basic Law envisions a decentralized system in which each agency determines its own risk classification and management practices[3]. The U.S. framework aligns closely with this philosophy. By empowering agencies to identify high-risk AI use cases tailored to their specific contexts, M-25-21 helps ensure that AI governance remains grounded in operational realities. At the same time, adopting M-25-21’s baseline practices, waiver safeguards, and disclosure requirements would provide consistency and public accountability across agencies. The combination of minimum risk management practices and transparent waiver use would encourage innovation while reassuring the public that any exceptions are justified, continuously monitored, and effectively controlled. Furthermore, embracing an approach that reflects emerging international consensus—particularly the emphasis on transparency in both U.S. and EU regimes—would position Taiwan to harmonize with global AI governance trends and strengthen its credibility in international markets. In contrast, the European Union’s AI Act predefines high-risk categories and mandates strict conformity assessments, CE Marking, and post-market monitoring[4]—an approach that is comprehensive but resource-intensive and may not suit all agencies equally. Australia’s ongoing discussions had been trending toward a similarly comprehensive model, but there has recently been backlash against this approach. Korea’s AI Basic Act[5] references high-risk AI only in broad terms and leaves most operational details undefined. M-25-21 strikes a middle ground, offering minimum yet concrete safeguards while preserving the flexibility agencies need to tailor governance to their specific domains. IV. Recommendations and Conclusion Based on this analysis, this research recommends that each agency designate a senior AI leader similar to a CAIO, maintain a public inventory of high-impact AI use cases, and publish summaries of waivers or determinations while safeguarding sensitive information. Agencies should also be encouraged to share AI resources and lessons learned to reduce duplication and strengthen governance maturity across government. Over time, these risk management practices can be refined in response to operational experience and evolving international standards. By adopting these principles, Taiwan can empower its agencies to innovate responsibly, protect citizens’ rights, and build public trust—ensuring that AI deployment across government remains both effective and aligned with global best practices. [1]〈政院通過「人工智慧基本法」草案 建構AI發展與應用良善環境 打造臺灣成為AI人工智慧島〉，行政院，https://www.ey.gov.tw/Page/9277F759E41CCD91/5d673d1e-f418-47dc-ab35-a06600f77f07（最後瀏覽日期︰2025/09/15）。 [2] United States Office of Management and Budget (OMB), M-25-21 Accelerating Federal Use of AI through Innovation, Governance, and Public Trust, https://www.whitehouse.gov/wp-content/uploads/2025/02/M-25-21-Accelerating-Federal-Use-of-AI-through-Innovation-Governance-and-Public-Trust.pdf (last visited Sept 15, 2025). [3] 蘇文彬，〈行政院通過AI基本法草案，將不設立AI專責機關〉，iThome，https://www.ithome.com.tw/news/170874（最後瀏覽日期︰2025/09/15）。 [4] Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 laying down harmonised rules on artificial intelligence and amending Regulations (EC) No 300/2008, (EU) No 167/2013, (EU) No 168/2013, (EU) 2018/858, (EU) 2018/1139 and (EU) 2019/2144 and Directives 2014/90/EU, (EU) 2016/797 and (EU) 2020/1828 (Artificial Intelligence Act), https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A32024R1689 (last visited Sept 15, 2025). [5] 인공지능발전과신뢰기반조성등에관한기본법안，https://www.law.go.kr/%EB%B2%95%EB%A0%B9/%EC%9D%B8%EA%B3%B5%EC%A7%80%EB%8A%A5%20%EB%B0 %9C%EC%A0%84%EA%B3%BC%20%EC%8B%A0%EB%A2%B0%20%EA%B8%B0%EB%B0%98%20%EC%A1%B0 %EC%84%B1%20%EB%93%B1%EC%97%90%20%EA%B4%80%ED%95%9C%20%EA%B8%B0%EB%B3%B8%EB%B2 %95/(20676,20250121) (last visited Sept 15, 2025).
Introduction to Taiwan’s Guidelines for Implementing Decentralized Elements in Medicinal Product Clinical TrialsIntroduction to Taiwan’s Guidelines for Implementing Decentralized Elements in Medicinal Product Clinical Trials 2023/12/15 The development of digital tools such as the internet, apps, and wearable devices have meant major breakthroughs for clinical trials. These advances have the potential to reduce the frequency of trial subject visits, accelerate research timelines, and lower the costs of drug development. The COVID-19 pandemic has further accelerated the use of digital tools, prompting many countries to adopt decentralized measures that enable trial subjects to participate in clinical trials regardless of their physical location. In step with the transition into the post-pandemic era, the Taiwan Food and Drug Administration (TFDA) issued the Guidelines for Implementing Decentralized Elements in Medicinal Product Clinical Trials in June, 2023[1]. The Guidelines are intended to cover a wide array of decentralized measures; they aim to increase trial subjects’ willingness to participate in trials, reduce the need for in-person visits to clinical trial sites, enhance real-time data acquisition during trials, and enable clinic sponsors and contract research organizations to process data remotely. I. Key Points of Taiwan’s Guidelines for Implementing Decentralized Elements in Medicinal Product Clinical Trials The Guidelines cover primarily the following matters: General considerations for implementing decentralized measures; trial subject recruitment and electronic informed consent; delivery and provision of investigational medicinal products; remote monitoring of trial subject safety; trial subject reporting of adverse events; remote data monitoring; and information systems and electronic data collection/processing/storage. 1. General Considerations for Implementing Decentralized Measures (1) During clinical trial execution, a reduction in trial subject in-person visits may present challenges to medical observation. It is recommended that home visits for any given trial subject be conducted by the principal investigator, sub-investigator, or a single, consistent delegated study nurse. (2) Sponsors must carefully evaluate all of the trial design’s decentralization measures to ensure data integrity. (3) Sponsors must conduct risk assessments for each individual trial, and must confirm the rationality of choosing decentralized measures. These decentralized measures must also be incorporated into the protocol. (4) When electronically collecting data, sponsors must ensure information system reliability and data security. Artificial intelligence may be considered for use in decentralized clinical trials; sponsors must carefully evaluate such systems, especially when they touch on determinations for critical data or strategies. (5) As the design of decentralized clinical trials is to ensure equal access to healthcare services, it must provide patients with a variety of ways to participate in clinical trials. (6) When implementing any decentralized measures, it is essential to ensure that the principal investigator and sponsor adhere to the Regulations for Good Clinical Practice and bear their respective responsibilities for the trial. (7) The use of decentralized measures must be stated in the regulatory application, and the Checklist of Decentralized Elements in Medicinal Product Clinical Trials must be included in the submission. 2. Subject Recruitment and Electronic Informed Consent (1) Trial subject recruitment through social media or established databases may only be implemented after the Institutional Review Board reviews and approves of the recruitment methods and content. (2) Must comply with the Principles for Recruiting Clinical Trial Subjects in medicinal product trials, the Personal Data Protection Act, and other regulations. (3) Regarding clinical trial subject informed consent done through digital software or devices, if it complies with Article 4, Paragraph 2 of the Electronic Signatures Act, that is, if the content can be displayed in its entirety and continues to be accessible for subsequent reference, then so long as the trial subject agrees to do so, the signature may be done via a tablet or other electronic device. The storage of signed electronic Informed Consent Forms (eICF) must align with the aforementioned Principles and meet the competent authority’s access requirements. 3. Delivery and Provision of Investigational Medicinal Products (1) The method of delivering and providing investigational medicinal products and whether trial subjects can use them on their own at home depends to a high degree on the investigational medicinal product’s administration route and safety profile. (2) When investigational medicinal products are delivered and provided through decentralized measures to trial subjects, this must be documented in the protocol. The process of delivering and providing said products must also be clearly stated in the informed consent form; only after being explained to a trial subject by the trial team, and after the trial subject’s consent is obtained, may such decentralized measures be used. (3) Investigational products prescribed by the principal investigator/sub-investigator must be reviewed by a delegated pharmacist to confirm that the investigational products’ specific items, dosage, duration, total quantity, and labeling align with the trial design. The pharmacist must also review each trial subject’s medication history, to ensure there are no medication-related issues; only then, and only in a manner that ensures the investigational product’s quality and the subject’s privacy, may delegated and specifically-trained trial personnel provide the investigational product to the subject. (4) Compliance with relevant regulations such as the Pharmaceutical Affairs Act, Pharmacists Act, Regulations on Good Practices for Drug Dispensation, and Regulations for Good Clinical Practice is required. 4. Remote Monitoring of Subject Safety (1) Decentralized trial designs involve trial subjects performing relatively large numbers of trial-related procedures at home. The principal investigator must delegate trained, qualified personnel to perform tasks such as collecting blood samples, administering investigational products, conducting safety monitoring, doing adverse event tracking, etc. (2) If trial subjects receive protocol-prescribed testing at nearby medical facilities or laboratories rather than at the original trial site, these locations must be authorized by the trial sponsor and must have relevant laboratory certification; only then may they collect or analyze samples. Such locations must provide detailed records to the principal investigator, to be archived in the trial master file. (3) The trial protocol and schedule must clearly specify which visits must be conducted at the trial site; which can be conducted via phone calls, video calls, or home visits; which tests must be performed at nearby laboratories; and whether trial subjects have multiple or single options at each visit. 5. Subject Reporting of Adverse Events (1) If the trial uses a digital platform to enhance adverse event reporting, trial subjects must be able to report adverse events through the digital platform, such as via a mobile phone app; that is, the principal investigator must be able to immediately access such adverse event information. (2) The principal investigator must handle such reports using risk-based assessment methods. The principal investigator must validate the adverse event reporting platform’s effectiveness, and must develop procedures to identify potential duplicate reports. 6. Remote Data Monitoring (1) If a sponsor chooses to implement remote monitoring, it must perform a reasonability assessment to confirm the appropriateness of such monitoring and establish a remote monitoring plan. (2) The monitoring plan must include monitoring strategies, monitoring personnel responsibilities, monitoring methods, rationale for such implementation, and critical data and processes that must be monitored. It must also generate comprehensive monitoring reports for audit purposes. (3) The sponsor is responsible for ensuring the implementation of remote monitoring, and must conduct risk assessments regarding the implementation process’ data protection and information confidentiality. 7. Information Systems and Electronic Data Collection, Processing, and Storage (1) In accordance with the Regulations for Good Clinical Practice, data recorded in clinical trials must be trustworthy, reliable, and verifiable. (2) It must be ensured that all organizations participating in the clinical trial have a full picture of the data flow. It is recommended that the trial protocol and trial-related documents include data flow diagrams and additional explanations. (3) Define the types and scopes of subject personal data that will be collected, and ensure that every step in the process properly protects their data in accordance with the Personal Data Protection Act. II. A Comparison with Decentralized Trial Regulations in Other Countries Denmark became the first country in the world to release regulatory measures on decentralized trials, issuing the “Danish Medicines Agency’s Guidance on the Implementation of Decentralized Elements in Clinical Trials with Medicinal Products” in September 2021[2]. In December 2022, the European Union as a whole released its “Recommendation Paper on Decentralized Elements in Clinical Trials”[3]. The United States issued the draft “Decentralized Clinical Trials for Drugs, Biological Products, and Devices” document in May 2023[4]. The comparison in Table 1 shows that Taiwan’s guidelines a relatively similar in structure to those of Denmark and the EU; the US guidelines also cover medical device clinical trials. Table 1: Summary of Decentralized Clinical Trial Guidelines in Taiwan, Denmark, the European Union as a whole, and the United States Taiwan Denmark European Union as a whole United States What do the guidelines apply to? Medicinal products Medicinal products Medicinal products Medicinal products and medical devices Trial subject recruitment and electronic informed consent Covers informed consent process; informed consent interview; digital information sheet; trial subject consent form signing; etc. Covers informed consent process; informed consent interview; trial subject consent form signing; etc. Covers informed consent process; informed consent interview; digital information sheet; trial subject consent form signing; etc. Covers informed consent process; informed consent interview; etc. Delivery and provision of investigational medicinal products Delegated, specifically-trained trial personnel deliver and provide investigational medicinal products. The investigator or delegated personnel deliver and provide investigational medicinal products. The investigator, delegated personnel, or a third-party, Good Distribution Practice-compliant logistics provider deliver and provide investigational medicinal products. The principal investigator, delegated personnel, or a distributor deliver and provide investigational products. Remote monitoring of trial subject safety Trial subjects may do return visits at trial sites, via phone calls, via video calls, or via home visits, and may undergo testing at nearby laboratories. Trial subjects may do return visits at trial sites, via phone calls, via video calls, or via home visits, and may undergo testing at nearby laboratories. Trial subjects may do return visits at trial sites, via phone calls, via video calls, or via home visits. Trial subjects may do return visits at trial sites, via phone calls, via video calls, or via home visits, and may undergo testing at nearby laboratories. Trial subject reporting of adverse events Trial subjects may self-report adverse events through a digital platform. Trial subjects may self-report adverse events through a digital platform. Trial subjects may self-report adverse events through a digital platform. Trial subjects may self-report adverse events through a digital platform. Remote data monitoring The sponsor may conduct remote data monitoring. The sponsor may conduct remote data monitoring. The sponsor may conduct remote data monitoring (not permitted in some countries). The sponsor may conduct remote data monitoring. Information systems and electronic data collection, processing, and storage The recorded data must be credible, reliable, and verifiable. Requires an information system that is validated, secure, and user-friendly. The recorded data must be credible, reliable, and verifiable. Must ensure data reliability, security, privacy, and confidentiality. III. Conclusion The implementation of decentralized clinical trials must be approached with careful assessment of risks and rationality, with trial subject safety, rights, and well-being as top priorities. Since Taiwan’s Guidelines for Implementing Decentralized Elements in Medicinal Product Clinical Trials were just announced in June of this year, the status of decentralized clinical trial implementation is still pending industry feedback to confirm feasibility. The overall goal is to enhance and optimize the clinical trial environment in Taiwan. Reference: [1] 衛生福利部食品藥物管理署，〈藥品臨床試驗執行分散式措施指引〉，2023/6/12，https://www.fda.gov.tw/TC/siteListContent.aspx?sid=9354&id=43548（最後瀏覽日：2023/11/2）。 [2] [DMA] DANISH MEDICINES AGENCY, The Danish Medicines Agency’s guidance on the Implementation of decentralised elements in clinical trials with medicinal products (2021),https://laegemiddelstyrelsen.dk/en/news/2021/guidance-on-the-implementation-of-decentralised-elements-in-clinical-trials-with-medicinal-products-is-now-available/ (last visited Nov. 2, 2023). [3] [HMA] HEADS OF MEDICINES AGENCIES, [EC] EUROPEAN COMMISSION & [EMA] EUROPEAN MEDICINES AGENCY, Recommendation paper on decentralised elements in clinical trials (2022),https://health.ec.europa.eu/latest-updates/recommendation-paper-decentralised-elements-clinical-trials-2022-12-14_en (last visited Nov. 2, 2023). [4] [US FDA] US FOOD AND DRUG ADMINISTRATION, Decentralized Clinical Trials for Drugs, Biological Products, and Devices (draft, 2023),https://www.fda.gov/regulatory-information/search-fda-guidance-documents/decentralized-clinical-trials-drugs-biological-products-and-devices (last visited Nov. 2, 2023).
Impact of Government Organizational Reform to Scientific Research Legal System and Response Thereto (1) – For Example, The Finnish Innovation Fund (“SITRA”)Impact of Government Organizational Reform to Scientific Research Legal System and Response Thereto (1) – For Example, The Finnish Innovation Fund (“SITRA”) I. Foreword We hereby aim to analyze and research the role played by The Finnish Innovation Fund (“Sitra”) in boosting the national innovation ability and propose the characteristics of its organization and operation which may afford to facilitate the deliberation on Taiwan’s legal system. Sitra is an independent organization which is used to reporting to the Finnish Parliament directly, dedicated to funding activities to boost sustainable development as its ultimate goal and oriented toward the needs for social change. As of 2004, it promoted the fixed-term program. Until 2012, it, in turn, primarily engaged in 3-year program for ecological sustainable development and enhancement of society in 2012. The former aimed at the sustainable use of natural resources to develop new structures and business models and to boost the development of a bioeconomy and low-carbon society, while the latter aimed to create a more well-being-oriented public administrative environment to upgrade various public sectors’ leadership and decision-making ability to introduce nationals’ opinion to policies and the potential of building new business models and venture capital businesses[1]. II. Standing and Operating Instrument of Sitra 1. Sitra Standing in Boosting of Finnish Innovation Policies (1) Positive Impact from Support of Innovation R&D Activities by Public Sector Utilization of public sector’s resources to facilitate and boost industrial innovation R&D ability is commonly applied in various countries in the world. Notwithstanding, the impact of the public sector’s investment of resources produced to the technical R&D and the entire society remains explorable[2]. Most studies still indicate positive impact, primarily as a result of the market failure. Some studies indicate that the impact of the public sector’s investment of resources may be observable at least from several points of view, including: 1. The direct output of the investment per se and the corresponding R&D investment potentially derived from investees; 2. R&D of outputs derived from the R&D investment, e.g., products, services and production methods, etc.; 3. direct impact derived from the R&D scope, e.g., development of a new business, or new business and service models, etc.; 4. impact to national and social economies, e.g., change of industrial structures and improvement of employment environment, etc. Most studies indicate that from the various points of view, the investment by public sector all produced positive impacts and, therefore, such investment is needed definitely[3]. The public sector may invest in R&D in diversified manners. Sitra invests in the “market” as an investor of corporate venture investment market, which plays a role different from the Finnish Funding Agency for Technology and Innovation (“Tekes”), which is more like a governmental subsidizer. Nevertheless, Finland’s characteristics reside in the combination of multiple funding and promotion models. Above all, due to the different behavior model, the role played by the former is also held different from those played by the general public sectors. This is why we choose the former as the subject to be studied herein. Data source: Jari Hyvärinen & Anna-Maija Rautiainen, Measuring additionality and systemic impacts of public research and development funding – the case of TEKES, FINLAND, RESEARCH EVALUATION, 16(3), 205, 206 (2007). Fig. 1 Phased Efforts of Resources Invested in R&D by Public Sector (2) Two Sided f Role Played by Sitra in Boosting of Finnish Innovation Policies Sitra has a very special position in Finland’s national innovation policies, as it not only helps successful implementation of the innovation policies but also acts an intermediary among the relevant entities. Sitra was founded in 1967 under supervision of the Bank of Finland before 1991, but was transformed into an independent foundation under the direction of the Finnish Parliament[4]. Though Sitra is a public foundation, its operation will not be intervened or restricted by the government. Sitra may initiate any innovation activities for its new organization or system, playing a role dedicated to funding technical R&D or promoting venture capital business. Meanwhile, Sitra also assumes some special function dedicated to decision-makers’ training and organizing decision-maker network to boost structural change. Therefore, Sitra may be identified as a special organization which may act flexibly and possess resources at the same time and, therefore, may initiate various innovation activities rapidly[5]. Sitra is authorized to boost the development of innovation activities in said flexible and characteristic manner in accordance with the Finland Innovation Fund Act (Laki Suomen itsenäisyyden juhlarahastosta). According to the Act, Finland established Sitra in 1967 and Sitra was under supervision of Bank of Finland (Article 1). Sitra was established in order to boost the stable growth of Finland’s economy via the national instrument’s support of R&D and education or other development instruments (Article 2). The policies which Sitra may adopt include loaning or funding, guarantee, marketable securities, participation in cooperative programs, partnership or equity investment (Article 3). If necessary, Sitra may collect the title of real estate or corporate shares (Article 7). Data source: Finnish innovation system, Research.fi, http://www.research.fi/en/innovationsystem.html (last visited Mar. 15, 2013). Fig. 2 Finnish Scientific Research Organization Chart Sitra's innovation role has been evolved through two changes. Specifically, Sitra was primarily dedicated to funding technical R&D among the public sectors in Finland, and the funding model applied by Sitra prior to the changes initiated the technical R&D promotion by Tekes, which was established in 1983. The first change of Sitra took place in 1987. After that, Sitra turned to focus on the business development and venture capital invested in technology business and led the venture capital investment. Meanwhile, it became a partner of private investment funds and thereby boosted the growth of venture capital investments in Finland in 1990. In 2000, the second change of Sitra took place and Sitra’s organization orientation was changed again. It achieved the new goal for structural change step by step by boosting the experimental social innovation activities. Sitra believed that it should play the role contributing to procedural change and reducing systematic obstacles, e.g., various organizational or institutional deadlocks[6]. Among the innovation policies boosted by the Finnish Government, the support of Start-Ups via governmental power has always been the most important one. Therefore, the Finnish Government is used to playing a positive role in the process of developing the venture capital investment market. In 1967, the Government established a venture capital company named Sponsor Oy with the support from Bank of Finland, and Sponsor Oy was privatized after 1983. Finland Government also established Kera Innovation Fund (now known as Finnvera[7]) in 1971, which was dedicated to boosting the booming of Start-Ups in Finland jointly with Finnish Industry Investment Ltd. (“FII”) established by the Government in 1994, and Sitra, so as to make the “innovation” become the main development force of the country[8] . Sitra plays a very important role in the foundation and development of venture capital market in Finland and is critical to the Finnish Venture Capital Association established in 1990. After Bank of Finland was under supervision of Finnish Parliament in 1991, Sitra became on the most important venture capital investors. Now, a large portion of private venture capital funds are provided by Sitra[9]. Since Sitra launched the new strategic program in 2004, it has turned to apply smaller sized strategic programs when investing young innovation companies, some of which involved venture capital investment. The mapping of young innovation entrepreneurs and angel investors started as of 1996[10]. In addition to being an important innovation R&D promoter in Finland, Sitra is also an excellent organization which is financially self-sufficient and tends to gain profit no less than that to be generated by a private enterprise. As an organization subordinated to the Finnish Parliament immediately, all of Sitra’s decisions are directly reported to the Parliament (public opinion). Chairman of Board, Board of Directors and supervisors of Sitra are all appointed by the Parliament directly[11]. Its working funds are generated from interest accruing from the Fund and investment income from the Fund, not tax revenue or budget prepared by the Government any longer. The total fund initially founded by Bank of Finland amounted to DEM100,000,000 (approximately EUR17,000,000), and was accumulated to DEM500,000,000 (approximately EUR84,000,000) from 1972 to 1992. After that, following the increase in market value, its nominal capital amounted to DEM1,400,000,000 (approximately EUR235,000,000) from 1993 to 2001. Obviously, Sitra generated high investment income. Until 2010, it has generated the investment income amounting to EUR697,000,000 . In fact, Sitra’s concern about venture capital investment is identified as one of the important changes in Finland's national technical R&D polices after 1990[13]. Sitra is used to funding businesses in three manners, i.e., direct investment in domestic stock, investment in Finnish venture capital funds, and investment in international venture capital funds, primarily in four industries, technology, life science, regional cooperation and small-sized & medium-sized starts-up. Meanwhile, it also invests in venture capital funds for high-tech industries actively. In addition to innovation technology companies, technical service providers are also its invested subjects[14]. 2. “Investment” Instrument Applied by Sitra to Boost Innovation Business The Starts-Up funding activity conducted by Sitra is named PreSeed Program, including INTRO investors’ mapping platform dedicated to mapping 450 angel investment funds and entrepreneurs, LIKSA engaged in working with Tekes to funding new companies no more than EUR40,000 for purchase of consultation services (a half thereof funded by Tekes, and the other half funded by Sitra in the form of loan convertible to shares), DIILI service[15] dedicated to providing entrepreneurs with professional sale consultation resources to integrate the innovation activity (product thereof) and the market to remedy the deficit in the new company’s ability to sell[16]. The investment subjects are stated as following. Sitra has three investment subjects, namely, corporate investments, fund investments and project funding. (1) Corporate investment Sitra will not “fund” enterprises directly or provide the enterprises with services without consideration (small-sized and medium-sized enterprises are aided by other competent authorities), but invest in the businesses which are held able to develop positive effects to the society, e.g., health promotion, social problem solutions, utilization of energy and effective utilization of natural resources. Notwithstanding, in order to seek fair rate of return, Sitra is dedicated to making the investment (in various enterprises) by its professional management and technology, products or competitiveness of services, and ranging from EUR300,000 to EUR1,000,000 to acquire 10-30% of the ownership of the enterprises, namely equity investment or convertible funding. Sitra requires its investees to value corporate social responsibility and actively participate in social activities. It usually holds the shares from 4 years to 10 years, during which period it will participate the corporate operation actively (e.g., appointment of directors)[17]. (2) Fund investments For fund investments[18], Sitra invests in more than 50 venture capital funds[19]. It invests in domestic venture capital fund market to promote the development of the market and help starts-up seek funding and create new business models, such as public-private partnerships. It invests in international venture capital funds to enhance the networking and solicit international funding, which may help Finnish enterprises access international trend information and adapt to the international market. (3) Project funding For project funding, Sitra provides the on-site information survey (supply of information and view critical to the program), analysis of business activities (analysis of future challenges and opportunities) and research & drafting of strategies (collection and integration of professional information and talents to help decision making), and commissioning of the program (to test new operating model by commissioning to deal with the challenge from social changes). Notwithstanding, please note that Sitra does not invest in academic study programs, research papers or business R&D programs[20]. (4) DIILI Investment Model Integrated With Investment Absorption A Start-Up usually will not lack technologies (usually, it starts business by virtue of some advanced technology) or foresighted philosophy when it is founded initially, while it often lacks the key to success, the marketing ability. Sitra DIILI is dedicated to providing the professional international marketing service to help starts-up gain profit successfully. Owing to the fact that starts-up are usually founded by R&D personnel or research-oriented technicians, who are not specialized in marketing and usually retains no sufficient fund to employ marketing professionals, DILLI is engaged in providing dedicated marketing talents. Now, it employs about 85 marketing professionals and seeks to become a start-up partner by investing technical services. Notwithstanding, in light of the characteristics of Sitra’s operation and profitability, some people indicate that it is more similar to a developer of an innovation system, rather than a neutral operator. Therefore, it is not unlikely to hinder some work development which might be less profitable (e.g., establishment of platform). Further, Sitra is used to developing some new investment projects or areas and then founding spin-off companies after developing the projects successfully. The way in which it operates seems to be non-compatible with the development of some industries which require permanent support from the public sector. The other issues, such as INTRO lacking transparency and Sitra's control over investment objectives likely to result in adverse choice, all arise from Sitra’s consideration to its own investment opportunities and profit at the same time of mapping. Therefore, some people consider that it should be necessary to move forward toward a more transparent structure or a non-income-oriented funding structure[21] . Given this, the influence of Sitra’s own income over upgrading of the national innovation ability when Sitra boosts starts-up to engage in innovation activities is always a concern remaining disputable in the Finnish innovation system. 3. Boosting of Balance in Regional Development and R&D Activities In order to fulfill the objectives under Lisbon Treaty and to enable EU to become the most competitive region in the world, European Commission claims technical R&D as one of its main policies. Among other things, under the circumstance that the entire R&D competitiveness upgrading policy is always progressing sluggishly, Finland, a country with a population of 5,300,000, accounting for 1.1% of the population of 27 EU member states, was identified as the country with the No. 1 innovation R&D ability in the world by World Economic Forum in 2005. Therefore, the way in which it promotes innovation R&D policies catches the public eyes. Some studies also found that the close relationship between R&D and regional development policies of Finland resulted in the integration of regional policies and innovation policies, which were separated from each other initially, after 1990[22]. Finland has clearly defined the plan to exploit the domestic natural resources and human resources in a balanced and effective manner after World War II. At the very beginning, it expanded the balance of human resources to low-developed regions, in consideration of the geographical politics, but in turn, it achieved national balanced development by meeting the needs for a welfare society and mitigation of the rural-urban divide as time went by. The Finnish innovation policies which may resort to technical policies retroactively initially drove the R&D in the manners including upgrading of education degree, founding of Science and Technology Policy Council and Sitra, establishment of Academy of Finland (1970) and establishment of the technical policy scheme, et al.. Among other things, people saw the role played by Sitra in Finland’s knowledge-intensive society policy again. From 1991 to 1995, the Finnish Government officially included the regional competitiveness into the important policies. The National Industrial Policy for Finland in 1993 adopted the strategy focusing on the development based on competitive strength in the regional industrial communities[23]. Also, some studies indicated that in consideration of Finland’s poor financial and natural resources, its national innovation system should concentrate the resources on the R&D objectives which meet the requirements about scale and essence. Therefore, the “Social Innovation, Social and Economic Energy Re-building Learning Society” program boosted by Sitra as the primary promoter in 2002 defined the social innovation as “the reform and action plan to enhance the regulations of social functions (law and administration), politics and organizational structure”, namely reform of the mentality and cultural ability via social structural changes that results in social economic changes ultimately. Notwithstanding, the productivity innovation activity still relies on the interaction between the enterprises and society. Irrelevant with the Finnish Government’s powerful direction in technical R&D activities, in fact, more than two-thirds (69.1%) of the R&D investment was launched by private enterprises and even one-thirds launched by a single enterprise (i.e., Nokia) in Finland. At the very beginning of 2000, due to the impact of globalization to Finland’s innovation and regional policies, a lot of R&D activities were emigrated to the territories outside Finland[24]. Multiple disadvantageous factors initiated the launch of national resources to R&D again. The most successful example about the integration of regional and innovation policies in Finland is the Centres of Expertise Programme (CEP) boosted by it as of 1990. Until 1994, there have been 22 centres of expertise distributed throughout Finland. The centres were dedicated to integrating local universities, research institutions and enterprise for co-growth. The program to be implemented from 2007 to 2013 planned 21 centres of expertise (13 groups), aiming to promote the corporate sectors’ cooperation and innovation activities. CEP integrated local, regional and national resources and then focused on the businesses designated to be developed[25]. [1] Sitra, http://www.sitra.fi/en (last visited Mar. 10, 2013). [2] Jari Hyvärinen & Anna-Maija Rautiainen, Measuring additionality and systemic impacts of public research and development funding – the case of TEKES, FINLAND, RESEARCH EVALUATION, 16(3), 205, 208 (2007). [3] id. at 206-214. [4] Charles Edquist, Tterttu Luukkonen & Markku Sotarauta, Broad-Based Innovation Policy, in EVALUATION OF THE FINNISH NATIONAL INNOVATION SYSTEM – FULL REPORT 11, 25 (Reinhilde Veugelers st al. eds., 2009). [5] id. [6] id. [7] Finnvera is a company specialized in funding Start-Ups, and its business lines include loaning, guarantee, venture capital investment and export credit guarantee, etc. It is a state-run enterprise and Export Credit Agency (ECA) in Finland. Finnvera, http://annualreport2012.finnvera.fi/en/about-finnvera/finnvera-in-brief/ (last visited Mar. 10, 2013). [8] Markku Maula, Gordon Murray & Mikko Jääskeläinen, MINISTRY OF TRADE AND INDUSTRY, Public Financing of Young Innovation Companies in Finland 32 (2006). [9] id. at 33. [10] id. at 41. [11] Sitra, http://www.sitra.fi/en (last visited Mar. 10, 2013). [12] Sitra, http://www.sitra.fi/en (last visited Mar. 10, 2013). [13] The other two were engaged in boosting the regional R&D center and industrial-academy cooperative center programs. Please see Gabriela von Blankenfeld-Enkvist, Malin Brännback, Riitta Söderlund & Marin Petrov, ORGANISATION FOR ECONOMIC CO-OPERATION AND DEVELOPMENT [OECD],OECD Case Study on Innovation: The Finnish Biotechnology Innovation System 15 (2004). [14] id. at20. [15] DIILI service provides sales expertise for SMEs, Sitra, http://www.sitra.fi/en/articles/2005/diili-service-provides-sales-expertise-smes-0 (last visited Mar. 10, 2013). [16] Maula, Murray & Jääskeläinen, supra note 8 at 41-42. [17] Corporate investments, Sitra, http://www.sitra.fi/en/corporate-investments (last visited Mar. 10, 2013). [18] Fund investments, Sitra, http://www.sitra.fi/en/fund-investments (last visited Mar. 10, 2013). [19] The venture capital funds referred to herein mean the pooled investment made by the owners of venture capital, while whether it exists in the form of fund or others is not discussed herein. [20] Project funding, Sitra, http://www.sitra.fi/en/project-funding (last visited Mar. 10, 2013). [21] Maula, Murray & Jääskeläinen, supra note 8 at 42. [22] Jussi S. Jauhiainen, Regional and Innovation Policies in Finland – Towards Convergence and/or Mismatch? REGIONAL STUDIES, 42(7), 1031, 1032-1033 (2008). [23] id. at 1036. [24] id. at 1038. [25] id. at 1038-1039.
Executive Yuan Promotes Free Economic Demonstration ZoneI.Background To promote more liberal and internationalized development of Taiwan economy, Premier of Executive Yuan approved the “Free Economic Demonstration Zone Plan” on April 26, 2013. Meanwhile, an Executive Yuan Working Group on Promotion of Economic Demonstration Zone is set up to accelerate the mapping out of the promotion programs as well as detailed action plans. The first phase of the Free Economic Demonstration Zone is to be officially initiated in July. According to the “Free Economic Demonstration Zone Plan”, the relevant laws and provisions regarding the flowing of human and financial capitals, and of logistics, will be loosen up to a great degree, based on the core ideas of liberalization, internationalization, and forwardness. Other related measures such as offering of lands and taxation would also be made, in order to attract capitals from both the inside and outside of the country. In addition, the Free Economic Demonstration Zone will first develop economic activities such as intelligent computing, international medicine services, value-added agriculture and cooperation among industries, to accelerate the transformation of the industrial structure of Taiwan. In order to construe an excellent environment for business of full liberalization and internationalization, the promotion strategies will be focused on “break-through of legal frameworks and innovations of management mechanisms”. II.Content of the Plan To accelerate the promotion process, the Free Economic Demonstration Zone will be conducted in two phases. The first phase is centered on the existing free trade port areas, including five ports and one airport, incorporated with the nature of “being inside the country border but outside the tariff zone”. All the industrial parks in the near counties and cities will also be integrated. The promotion will be set out simultaneously in the north, middle and south of Taiwan. The effects of the promotion are expected to be magnified by fully utilizing the resources and the unique characters of industries of each region. Moreover, the promulgation of a special legislation on the Free Economic Demonstration Zone would be facilitated in the future. After this special legislation is passed, the set-ups of demonstration zones can be applied by authorities either of central or of local government and the related promotion works of the second phase will be unfolded immediately. According to the Executive Yuan, the Free Economic Demonstration Zone will be beneficial in terms of creating positive conditions for Taiwan to participate in regional trade organizations and attract both local and foreign investment, injecting new movement into the economic growth of Taiwan. III.Recent Development In addition, on August 8, 2013, relevant discussions on “Furtherance Plan for Free Economic Demonstration Zone Phase One” are further unfolded in the Executive Yuan conference. In addition, the Premier also indicates, that the furtherance of the Free Economic Demonstration Zone (hereafter: FEDZ) is divided into two phases. The first phase starts from the moment that the Plan is approved till the related special legislation is passed and promulgated. In this phase, the relevant tasks can be achieved through the ways of promulgation of administrative orders. On the other hand, the tasks concerning taxation benefits and other parts that involve legislation will not able to be initiated till the second phase of the Plan. For those tasks, the Council for Economic Planning and Development is asked to complete the drafting of this special legislation and related procedures for registering it into the Executive Yuan, together with the Ministry of Economic Affairs and other concerned agencies, in the hope that the related legislation works of the Executive Yuan can be completed before the end of this year. In respect of “Furtherance Plan for Free Economic Demonstration Zone Phase One”, Premier Jiang further points out, that FEDZ is a model incorporates the concept of “being inside the country border but outside the tariff zone” and the idea of “combining the stores upfront and the factories behind, outsourcing manufactures”. In this way, the hinterland of a port can be expended and magnified effects to be achieved through using the resources provided by the factory in behind. Under this pattern, the expansion effects that cities and counties such as New Taipei City and Changhua Country fight for, can be further extended by this concept of “factories in the back”. As for Port of Anping, over which Tainan City government has proactively fought for, can be listed as a demonstration zone once the Executive Yuan approved it as free trade port zone. In the future, other places that are with forward-looking industry and suitable can still be enlisted. Premier Jiang further expresses that, there are four demonstration industries in the first phase, including intelligent computing, international medicine services, value-added agriculture and cooperation among industries. Yet, he also points out that the demonstration of liberalized economy is a concept of “4+N”. It means that the demonstration will not be limited to the scope of these four industries. Other industries that match up with the idea of liberalization, internationalization and foresight can all be incorporated into FEDZ through continuing examination. Moreover, Premier Jiang later mentions on August 14th, that FEDZ is a crucial task for the government at this moment. He thus requests the Ministry of Economic Affairs, Ministry of Transportation and Communications, Ministry of Health and Welfare, and the Council of Agriculture, to enhance the training and service quality of staffers of the single service window of furtherance of FEDZ. Moreover, Premier Jiang additionally indicated in November, that the scope of the FEDZ will include Pingtung Agricultural Biotechnology Park and Kaohsiung Free Trade Port Area. The combination of the two will facilitate adding value to the agriculture in Taiwan and put momentum into quality agriculture, making the high-quality agricultural products of our country being sold to all over the world with swift logistic services. Premier Jiang also mentioned, that in order to avoid Taiwan being marginalized amid regional integrations of global economies, the government is facilitating industries of potentials by proactively promote the FEDZ. The current approach is to expend the original free trade port area with legislative bases, creating the demonstration zones of free economy by combing original establishments such as Pingtung Agricultural Biotechnology Park. If this approach and system is proved feasible, the next step would be promoting it to island-wide, making the whole nation open-up. IV.Conclusion In the past decade, the economic development in Taiwan, compared to neighboring economic zones such as Hong Kong, Korean or Singapore, was indeed stagnant. It is thus a positive move for the government to put great efforts in promoting FEDZ, in the hope that the liberalization and internationalization of the economy of this country can therefore be significantly improved. Yet, some commentators are of slightly more skeptical opinions, reminding that in terms of the tax relaxation in the Plan, similar approach was already taken by the government before, which did not lead to the expected outcome. In sum, it still remains as a continuing task for us and for the administration as well, to ponder on how Taiwan can find out its own unique strength in the face of global competition. How we can attract more international partners, to create mutual economic benefits. The FEDZ is undoubtedly a first step. Nevertheless, challenges are still ahead of the government, as to how to take many more steps in the future, in order to make Taiwan to march on the stage of the world again.
- Impact of Government Organizational Reform to Scientific Research Legal System and Response Thereto (1) – For Example, The Finnish Innovation Fund (“SITRA”)
- The Demand of Intellectual Property Management for Taiwanese Enterprises
- Impact of Government Organizational Reform to Research Legal System and Response Thereto (2) – Observation of the Swiss Research Innovation System
- Blockchain in Intellectual Property Protection
- Recent Federal Decisions and Emerging Trends in U.S. Defend Trade Secrets Act Litigation
- Copyright Ownership for Outputs by Artificial Intelligence
- The effective and innovative way to use the spectrum: focus on the development of the "interleaved/white space"
- Impact of Government Organizational Reform to Research Legal System and Response Thereto (2) – Observation of the Swiss Research Innovation System