> Science & Technology R&D Law
Back
Previous Strengthening Taiwan’s Pharmaceutical Resilience: Legal Reflections from the European Union’s Critical Medicines Act Next Shifting AI Governance in East Asia: AI Legislative Progress in Japan, South Korea and Taiwan

A Preferred Model for Taiwan’s agency level AI risk categorization and management: A Cross-Jurisdictional Perspective

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).

※A Preferred Model for Taiwan’s agency level AI risk categorization and management: A Cross-Jurisdictional Perspective,STLI, https://stli.iii.org.tw/en/article-detail.aspx?no=55&tp=2&i=168&d=9402 (Date:2026/06/14)
Quote this paper

Chen, Chen
 Project Manager

Release： 2025/10
Tag
Read with phone
Back
Previous Strengthening Taiwan’s Pharmaceutical Resilience: Legal Reflections from the European Union’s Critical Medicines Act Next Shifting AI Governance in East Asia: AI Legislative Progress in Japan, South Korea and Taiwan
You may be interested
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).

The Institutionalization of the Taiwan Personal Data Protection Committee - Triumph of Digital Constitutionalism: A Legal Positivism Analysis

The Institutionalization of the Taiwan Personal Data Protection Committee - Triumph of Digital Constitutionalism: A Legal Positivism Analysis 2023/07/13 The Legislative Yuan recently passed an amendment to the Taiwan Personal Data Protection Act, which resulted in the institutionalization of the Taiwan Personal Data Protection Commission (hereunder the “PDPC”)[1]. This article aims to analyze the significance of this institutionalization from three different perspectives: legal positivism, digital constitutionalism, and Millian liberalism. By examining these frameworks, we can better understand the constitutional essence of sovereignty, the power dynamics among individuals, businesses, and governments, and the paradox of freedom that the PDPC addresses through governance and trust. I.Three Layers of Significance 1.Legal Positivism The institutionalization of the PDPC fully demonstrates the constitutional essence of sovereignty in the hands of citizens. Legal positivism emphasizes the importance of recognizing and obeying (the sovereign, of which it is obeyed by all but does not itself obey to anyone else, as Austin claims) laws that are enacted by legitimate authorities[2]. In this context, the institutionalization of the PDPC signifies the recognition of citizens' rights to control their personal data and the acknowledgment of the sovereign in protecting their privacy. It underscores the idea that the power to govern personal data rests with the individuals themselves, reinforcing the principles of legal positivism regarding sovereign Moreover, legal positivism recognizes the authority of the state in creating and enforcing laws. The institutionalization of the PDPC as a specialized commission with the power to regulate and enforce personal data protection laws represents the state's recognition of the need to address the challenges posed by the digital age. By investing the PDPC with the authority to oversee the proper handling and use of personal data, the state acknowledges its responsibility to protect the rights and interests of its citizens. 2.Digital Constitutionalism The institutionalization of the PDPC also rebalances the power structure among individuals, businesses, and governments in the digital realm[3]. Digital constitutionalism refers to the principles and norms that govern the relationship between individuals and the digital sphere, ensuring the protection of rights and liberties[4]. With the rise of technology and the increasing collection and use of personal data, individuals often find themselves at a disadvantage compared to powerful entities such as corporations and governments[5]. However, the PDPC acts as a regulatory body that safeguards individuals' interests, rectifying the power imbalances and promoting digital constitutionalism. By establishing clear rules and regulations regarding the collection, use, and transfer of personal data, the PDPC may set a framework that ensures the protection of individuals' privacy and data rights. It may enforce accountability among businesses and governments, holding them responsible for their data practices and creating a level playing field where individuals have a say in how their personal data is handled. 3.Millian Liberalism The need for the institutionalization of the PDPC embodies the paradox of freedom, as raised in John Stuart Mill’s “On Liberty”[6], where Mill recognizes that absolute freedom can lead to the infringement of others' rights and well-being. In this context, the institutionalization of the PDPC acknowledges the necessity of governance to mitigate the risks associated with personal data protection. In the digital age, the vast amount of personal data collected and processed by various entities raises concerns about privacy, security, and potential misuse. The institutionalization of the PDPC represents a commitment to address these concerns through responsible governance. By setting up rules, regulations, and enforcement mechanisms, the PDPC ensures that individuals' freedoms are preserved without compromising the rights and privacy of others. It strikes a delicate balance between individual autonomy and the broader social interest, shedding light on the paradox of freedom. II.Legal Positivism: Function and Authority of the PDPC 1.John Austin's Concept of Legal Positivism: Sovereignty, Punishment, Order To understand the function and authority of the PDPC, we turn to John Austin's concept of legal positivism. Austin posited that laws are commands issued by a sovereign authority and backed by sanctions[7]. Sovereignty entails the power to make and enforce laws within a given jurisdiction. In the case of the PDPC, its institutionalization by the Legislative Yuan reflects the recognition of its authority to create and enforce regulations concerning personal data protection. The PDPC, as an independent and specialized committee, possesses the necessary jurisdiction and competence to ensure compliance with the law, administer punishments for violations, and maintain order in the realm of personal data protection. 2.Dire Need for the Institutionalization of the PDPC There has been a dire need for the establishment of the PDPC following the Constitutional Court's decision in August 2022, holding that the government needed to establish a specific agency in charge of personal data-related issues[8]. This need reflects John Austin's concept of legal positivism, as it highlights the demand for a legitimate and authoritative body to regulate and oversee personal data protection. The PDPC's institutionalization serves as a response to the growing concerns surrounding data privacy, security breaches, and the increasing reliance on digital platforms. It signifies the de facto recognition of the need for a dedicated institution to safeguard the individual’s personal data rights, reinforcing the principles of legal positivism. Furthermore, the institutionalization of the PDPC demonstrates the responsiveness of the legislative branch to the evolving challenges posed by the digital age. The amendment to the Taiwan Personal Data Protection Act and the subsequent institutionalization of the PDPC are the outcomes of a democratic process, reflecting the will of the people and their desire for enhanced data protection measures. It signifies a commitment to uphold the rule of law and ensure the protection of citizens' rights in the face of emerging technologies and their impact on privacy. 3.Authority to Define Cross-Border Transfer of Personal Data Upon the establishment of the PDPC, it's authority to define what constitutes a cross-border transfer of personal data under Article 21 of the Personal Data Protection Act will then align with John Austin's theory on order. According to Austin, laws bring about order by regulating behavior and ensuring predictability in society. By granting the PDPC the power to determine cross-border data transfers, the legal framework brings clarity and consistency to the process. This promotes order by establishing clear guidelines and standards, reducing uncertainty, and enhancing the protection of personal data in the context of international data transfers. The PDPC's authority in this regard reflects the recognition of the need to regulate and monitor the cross-border transfer of personal data to protect individuals' privacy and prevent unauthorized use or abuse of their information. It ensures that the transfer of personal data across borders adheres to legal and ethical standards, contributing to the institutionalization of a comprehensive framework for cross-border data transfer. III.Conclusion In conclusion, the institutionalization of the Taiwan Personal Data Protection Committee represents the convergence of legal positivism, digital constitutionalism, and Millian liberalism. It signifies the recognition of citizens' sovereignty over their personal data, rebalances power dynamics in the digital realm, and addresses the paradox of freedom through responsible governance. By analyzing the PDPC's function and authority in the context of legal positivism, we understand its role as a regulatory body to maintain order and uphold the principles of legal positivism. The institutionalization of the PDPC serves as a milestone in Taiwan's commitment to protect individuals' personal data and safeguard the digital rights. In essence, the institutionalization of the Taiwan Personal Data Protection Committee represents a triumph of digital constitutionalism, where individuals' rights and interests are safeguarded, and power imbalances are rectified. It also embodies the recognition of the paradox of freedom and the need for responsible governance in the digital age in Taiwan. Reference: [1] Lin Ching-yin & Evelyn Yang, Bill to establish data protection agency clears legislative floor, CNA English News, FOCUS TAIWAN, May 16, 2023, https://focustaiwan.tw/society/202305160014 (last visited, July 13, 2023). [2] Legal positivism, Stanford Encyclopedia of Philosophy, https://plato.stanford.edu/entries/legal-positivism/?utm_source=fbia (last visited July 13, 2023). [3] Edoardo Celeste, Digital constitutionalism: how fundamental rights are turning digital, (2023): 13-36, https://doras.dcu.ie/28151/1/2023_Celeste_DIGITAL%20CONSTITUTIONALISM_%20HOW%20FUNDAMENTAL%20RIGHTS%20ARE%20TURNING%20DIGITAL.pdf (last visited July 3, 2023). [4] GIOVANNI DE GREGORIO, DIGITAL CONSTITUTIONALISM IN EUROPE: REFRAMING RIGHTS AND POWERS IN THE ALGORITHMIC SOCIETY 218 (2022). [5] Celeste Edoardo, Digital constitutionalism: how fundamental rights are turning digital (2023), https://doras.dcu.ie/28151/1/2023_Celeste_DIGITAL%20CONSTITUTIONALISM_%20HOW%20FUNDAMENTAL%20RIGHTS%20ARE%20TURNING%20DIGITAL.pdf (last visited July 13, 2023). [6]JOHN STUART MILL,On Liberty (1859), https://openlibrary-repo.ecampusontario.ca/jspui/bitstream/123456789/1310/1/On-Liberty-1645644599.pdf (last visited July 13, 2023). [7] Legal positivism, Stanford Encyclopedia of Philosophy, https://plato.stanford.edu/entries/legal-positivism/?utm_source=fbia (last visited July 13, 2023). [8] Lin Ching-yin & Evelyn Yang, Bill to establish data protection agency clears legislative floor, CNA English News, FOCUS TAIWAN, May 16, 2023, https://focustaiwan.tw/society/202305160014 (last visited, July 13, 2023).

Finland’s Technology Innovation System

I. Introduction 　　When, Finland, this country comes to our minds, it is quite easy for us to associate with the prestigious cell-phone company “NOKIA”, and its unbeatable high technology communication industry. However, following the change of entire cell-phone industry, the rise of smart phone not only has an influence upon people’s communication and interaction, but also makes Finland, once monopolized the whole cell-phone industry, feel the threat and challenge coming from other new competitors in the smart phone industry. However, even though Finland’s cell-phone industry has encountered frustrations in recent years in global markets, the Finland government still poured many funds into the area of technology and innovation, and brought up the birth of “Angry Birds”, one of the most popular smart phone games in the world. The Finland government still keeps the tradition to encourage R&D, and wishes Finland’s industries could re-gain new energy and power on technology innovation, and indirectly reach another new competitive level. 　　According to the Statistics Finland, 46% Finland’s enterprises took innovative actions upon product manufacturing and the process of R&D during 2008-2010; also, the promotion of those actions not merely existed in enterprises, but directly continued to the aspect of marketing and manufacturing. No matter on product manufacturing, the process of R&D, the pattern of organization or product marketing, we can observe that enterprises or organizations make contributions upon innovative activities in different levels or procedures. In the assignment of Finland’s R&D budgets in 2012, which amounted to 200 million Euros, universities were assigned by 58 million Euros and occupied 29% R&D budgets. The Finland Tekes was assigned by 55 million Euros, and roughly occupied 27.5% R&D budgets. The Academy of Finland (AOF) was assigned by 32 million Euros, and occupied 16% R&D budges. The government’s sectors were assigned by 3 million Euros, and occupied 15.2% R&D budgets. Other technology R&D expenses were 2.1 million Euros, and roughly occupied 10.5% R&D. The affiliated teaching hospitals in universities were assigned by 0.36 million Euros, and occupied 1.8% R&D budgets. In this way, observing the information above, concerning the promotion of technology, the Finland government not only puts more focus upon R&D innovation, but also pays much attention on education quality of universities, and subsidizes various R&D activities. As to the Finland government’s assignment of budges, it can be referred to the chart below. 　　As a result of the fact that Finland promotes industries’ innovative activities, it not only made Finland win the first position in “Growth Competitiveness Index” published by the World Economic Forum (WEF) during 2000-2006, but also located the fourth position in 142 national economy in “The Global Competitiveness Report” published by WEF, preceded only by Swiss, Singapore and Sweden, even though facing unstable global economic situations and the European debt crisis. Hence, observing the reasons why Finland’s industries have so strong innovative power, it seems to be related to the Finland’s national technology administrative system, and is worthy to be researched. II. The Recent Situation of Finland’s Technology Administrative System A. Preface 　　Finland’s administrative system is semi-presidentialism, and its executive power is shared by the president and the Prime Minister; as to its legislative power, is shared by the Congress and the president. The president is the Finland’s leader, and he/she is elected by the Electoral College, and the Prime Minister is elected by the Congress members, and then appointed by the president. To sum up, comparing to the power owned by the Prime Minister and the president in the Finland’s administrative system, the Prime Minister has more power upon executive power. So, actually, Finland can be said that it is a semi-predisnetialism country, but trends to a cabinet system. 　　Finland technology administrative system can be divided into four parts, and the main agency in each part, based upon its authority, coordinates and cooperates with making, subsidizing, executing of Finland’s technology policies. The first part is the policy-making, and it is composed of the Congress, the Cabinet and the Research and Innovation Council; the second part is policy management and supervision, and it is leaded by the Ministry of Education and Culture, the Ministry of Employment and the Economy, and other Ministries; the third part is science program management and subsidy, and it is composed of the Academy of Finland (AOF), the National Technology Agency (Tekes), and the Finnish National Fund Research and Development (SITRA); the fourth part is policy-executing, and it is composed of universities, polytechnics, public-owned research institutions, private enterprises, and private research institutions. Concerning the framework of Finland’s technology administrative, it can be referred to below. B. The Agency of Finland’s Technology Policy Making and Management (A) The Agency of Finland’s Technology Policy Making 　　Finland’s technology policies are mainly made by the cabinet, and it means that the cabinet has responsibilities for the master plan, coordinated operation and fund-assignment of national technology policies. The cabinet has two councils, and those are the Economic Council and the Research and Innovation Council, and both of them are chaired by the Prime Minister. The Research and Innovation Council is reshuffled by the Science and Technology Policy Council (STPC) in 1978, and it changed name to the Research and Innovation Council in Jan. 2009. The major duties of the Research and Innovation Council include the assessment of country’s development, deals with the affairs regarding science, technology, innovative policy, human resource, and provides the government with aforementioned schedules and plans, deals with fund-assignment concerning public research development and innovative research, coordinates with all government’s activities upon the area of science, technology, and innovative policy, and executes the government’s other missions. 　　The Research and Innovation Council is an integration unit for Finland’s national technology policies, and it originally is a consulting agency between the cabinet and Ministries. However, in the actual operation, its scope of authority has already covered coordination function, and turns to direct to make all kinds of policies related to national science technology development. In addition, the consulting suggestions related to national scientific development policies made by the Research and Innovation Council for the cabinet and the heads of Ministries, the conclusion has to be made as a “Key Policy Report” in every three year. The Report has included “Science, Technology, Innovation” in 2006, “Review 2008” in 2008, and the newest “Research and Innovation Policy Guidelines for 2011-2015” in 2010. 　　Regarding the formation and duration of the Research and Innovation Council, its duration follows the government term. As for its formation, the Prime Minister is a chairman of the Research and Innovation Council, and the membership consists of the Minister of Education and Science, the Minister of Economy, the Minister of Finance and a maximum of six other ministers appointed by the Government. In addition to the Ministerial members, the Council shall comprise ten other members appointed by the Government for the parliamentary term. The Members must comprehensively represent expertise in research and innovation. The structure of Council includes the Council Secretariat, the Administrative Assistant, the Science and Education Subcommittee, and the Technology and Innovation Subcommittee. The Council has the Science and Education Subcommittee and the Technology and Innovation Subcommittee with preparatory tasks. There are chaired by the Ministry of Education and Science and by the Minister of Economy, respectively. The Council’s Secretariat consists of one full-time Secretary General and two full-time Chief Planning Officers. The clerical tasks are taken care of at the Ministry of Education and Culture. (B) The Agency of Finland’s Technology Policy Management 　　The Ministries mainly take the responsibility for Finland’s technology policy management, which includes the Ministry of Education and Culture, the Ministry of Employment and Economy, the Ministry of Social Affairs and Health, the Ministry of Agriculture and Forestry, the Ministry of Defense, the Ministry of Transport and Communication, the Ministry of Environment, the Ministry of Financial, and the Ministry of Justice. In the aforementioned Ministries, the Ministry of Education and Culture and the Ministry of Employment and Economy are mainly responsible for Finland national scientific technology development, and take charge of national scientific policy and national technical policy, respectively. The goal of national scientific policy is to promote fundamental scientific research and to build up related scientific infrastructures; at the same time, the authority of the Ministry of Education and Culture covers education and training, research infrastructures, fundamental research, applied research, technology development, and commercialization. The main direction of Finland’s national scientific policy is to make sure that scientific technology and innovative activities can be motivated aggressively in universities, and its objects are, first, to raise research funds and maintain research development in a specific ratio; second, to make sure that no matter on R&D institutions or R&D training, it will reach fundamental level upon funding or environment; third, to provide a research network for Finland, European Union and global research; fourth, to support the research related to industries or services based upon knowledge-innovation; fifth, to strengthen the cooperation between research initiators and users, and spread R&D results to find out the values of commercialization, and then create a new technology industry; sixth, to analyze the performance of national R&D system. 　　As for the Ministry of Employment and Economy, its major duties not only include labor, energy, regional development, marketing and consumer policy, but also takes responsibilities for Finland’s industry and technical policies, and provides industries and enterprises with a well development environment upon technology R&D. The business scope of the Ministry of Employment and Economy puts more focus on actual application of R&D results, it covers applied research of scientific technology, technology development, commercialization, and so on. The direction of Finland’s national technology policy is to strengthen the ability and creativity of industries’ technology development, and its objects are, first, to develop the new horizons of knowledge with national innovation system, and to provide knowledge-oriented products and services; second, to promote the efficiency of the government R&D funds; third, to provide cross-country R&D research networks, and support the priorities of technology policy by strengthening bilateral or multilateral cooperation; fourth, to raise and to broaden the efficiency of research discovery; fifth, to promote the regional development by technology; sixth, to evaluate the performance of technology policy; seventh, to increase the influence of R&D on technological change, innovation and society; eighth, to make sure that technology fundamental structure, national quality policy and technology safety system will be up to international standards. (C) The Agency of Finland’s Technology Policy Management and Subsidy 　　As to the agency of Finland’s technology policy management and subsidy, it is composed of the Academy of Finland (AOF), the National Technology Agency (Tekes), and the Finnish National Fund Research and Development (SITRA). The fund of AOF comes from the Ministry of Education and Culture; the fund of Tekes comes from the Ministry of Employment and Economy, and the fund of SITRA comes from independent public fund supervised by the Finland’s Congress. (D) The Agency of Finland’s Technology Plan Execution 　　As to the agency of Finland’s technology plan execution, it mainly belongs to the universities under Ministries, polytechnics, national technology research institutions, and other related research institutions. Under the Ministry of Education and Culture, the technology plans are executed by 16 universities, 25 polytechnics, and the Research Institute for the Language of Finland; under the Ministry of Employment and Economy, the technology plans are executed by the Technical Research Centre of Finland (VTT), the Geological Survey of Finnish, the National Consumer Research Centre; under the Ministry of Social Affairs and Health, the technology plans are executed by the National Institute for Health and Welfare, the Finnish Institute of Occupational Health, and University Central Hospitals; under the Ministry of Agriculture and Forestry, the technology plans are executed by the Finnish Forest Research Institute (Metla), the Finnish Geodetic Institute, and the Finnish Game and Fisheries Research Institute (RKTL); under the Ministry of Defense, the technology plans are executed by the Finnish Defense Forces’ Technical Research Centre (Pvtt); under the Ministry of Transport and Communications, the technology plans are executed by the Finnish Meteorological Institute; under the Ministry of Environment, the technology plans are executed by the Finnish Environment Institute (SYKE); under the Ministry of Financial, the technology plans are executed by the Government Institute for Economic Research (VATT). At last, under the Ministry of Justice, the technology plans are executed by the National Research Institute of Legal Policy.

Analyzing the Framwork of the Regulation「Act For The Development of Biotech And New Pharmaceuticals Industry」in Taiwan

Taiwan Government passed The「Act for the Development of Biotech and New Pharmaceuticals Industry」for supporting the biopharmaceutical industry. The purpose of the Act is solely for biopharmaceutical industry, and building the leading economic force in Taiwan. To fulfill this goal, the Act has enacted regulations concerning funding, taxation and recruitment especially for the biopharmaceutical industry. The Act has been seen as the recent important law in the arena of upgrading industry regulation on the island. It is also a rare case where single legislation took place for particular industry. After the Act came into force, the government has promulgated further regulations to supplement the Act, including Guidance for MOEA-Approved Biotech and New Pharmaceuticals Company Issuing Stock Certificate, Deductions on Investments in R&D and Personnel Training of Biotech and New Pharmaceuticals Company, Guidance for Deduction Applicable to Shareholders of Profit-Seeking Enterprises -Biotech and New Pharmaceuticals Company etc. The following discussions are going to introduce the Act along with related incentive measures from an integrated standpoint. 1 、 Scope of Application According to Article 3 of the Act, 「Biotech and New Pharmaceuticals Industry」 refers to the industry that deals in New Rugs and High-risk Medical devices used by human beings, animals, and plants; 「Biotech and New Pharmaceuticals Company」 refers to a company in the Biotech and New Pharmaceuticals Industry that is organized and incorporated in accordance with the Company Act and engages in the research, development, and manufacture of new drugs and high-risk medical devices. Thus, the Act applies to company that conducts research and manufacture product in new drug or high-risk medical devices for human and animal use. Furthermore, to become a Biotech and New Pharmaceuticals Company stipulated in the Act, the Company must receive letter of approval to establish as a Biotech and New Pharmaceuticals Company valid for five years. Consequently, company must submit application to the authority for approval by meeting the following requirements: (1) Companies that conduct any R&D activities or clinical trials must receive permission, product registration, or proof of manufacture for such activities from a competent authority. However, for those conducted these activities outside the country will not apply. (2) When applied for funding for the previous year or in the same year, the expense on R&D in the previous year exceeds 5% of the total net revenue within the same year; or the expenses exceeds 10% of the total capital of the company. (3) Hired at least five R&D personnel majored in biotechnology. For New Drug and High-Risk Medical Device are confined in specific areas. New Drug provided in the Act refers to a drug that has a new ingredient, a new therapeutic effect or a new administration method as verified by the central competent authorities. And High-Risk Medical Device refers to a type of Class III medical devices implanted into human bodies as verified by the central competent authorities. Therefore, generic drug, raw materials, unimplanted medical device, and medical device are not qualified as type III, are all not within the scope of the Act and are not the subject matter the Act intends to reward. 2 、 Tax Benefits Article 5, 6 and 7 provided in the Act has followed the footsteps of Article 6 and 8 stipulated of the Statute, amending the rules tailored to the biopharmaceutical industry, and provided tax benefits to various entities as 「Biotech and New Pharmaceuticals Company」, 「Investors of Biotech and New Pharmaceuticals Industry」, 「Professionals and Technology Investors」. (1) Biotech and New Pharmaceuticals Company In an effort to advance the biopharmaceutical industry, alleviate financial burden of the companies and strengthen their R&D capacity. The Act has provided favorable incentive measures in the sector of R&D and personnel training. According to Article 5: 「For the purpose of promoting the Biotech and New Pharmaceuticals Industry, a Biotech and New Pharmaceuticals Company may, for a period of five years from the time it is subject to profit-seeking enterprise income tax payable, enjoy a reduction in its corporate income tax payable, for up to 35% of the total funds invested in research and development (R&D) and personnel training each year.」 Consequently, company could benefit through tax deduction and relieve from the stress of business operation. Moreover, in supporting Biotech and New Pharmaceutical Company to proceed in R&D and personnel training activities, the Act has set out rewards for those participate in ongoing R&D and training activities. As Article 5 provided that」 If the R&D expenditure of a particular year exceeds the average R&D expenditure of the previous two years, or if the personnel training expenditure of a particular year exceeds the average personnel training expenditure of the pervious two years, 50% of the exceed amount in excess of the average may be used to credit against the amount of profit-seeking enterprise income tax payable. 「However, the total amount of investment credited against by the payable corporate income tax in each year shall not exceed 50% of the amount of profit-seeking enterprise income tax payable by a Biotech and New Pharmaceuticals Company in a year, yet this restriction shall not apply to the amount to be offset in the last year of the aforementioned five-year period. Lastly, Article 5 of the Act shall not apply to Biotech and New Pharmaceutical Company that set up headquarters or branches outside of Taiwan. Therefore, to be qualified for tax deduction on R&D and personnel training, the headquarters or branches of the company must be located in Taiwan. (2) Investors of Biotech and New Pharmaceuticals Company To raise funding, expand business development, and attract investor continuing making investments, Article 6 of the Act has stated that 「In order to encourage the establishment or expansion of Biotech and New Pharmaceuticals Companies, a profit-seeking enterprise that subscribes for the stock issued by a Biotech and New Pharmaceuticals Company at the time of the latter's establishment or subsequent expansion; and has been a registered shareholder of the Biotech and New Pharmaceuticals Company for a period of 3 years or more, may, for a period of five years from the time it is subject to corporate income tax, enjoy a reduction in its profit-seeking enterprise income tax payable for up to 20% of the total amount of the price paid for the subscription of shares in such Biotech and New Pharmaceuticals Company.」 Yet 「If the afore-mentioned profit-seeking enterprise is a venture capital company (「VC」), such VC corporate shareholders may, for a period of five years from the fourth anniversary year of the date on which the VC becomes a registered shareholder of the subject Biotech and New Pharmaceuticals Company, enjoy a reduction in their profit-seeking enterprise income tax payable based on the total deductible amount enjoyed by the VC under Paragraph 1 hereof and the shareholders' respective shareholdings in the VC.」 The government enacted this regulation to encourage corporations and VC to invest in biotech and new pharmaceutical company, and thus provide corporate shareholders with 20% of profit-seeking enterprise income tax payable deduction, and provide VC corporate shareholders tax deduction that proportion to its shareholdings in the VC. (3) Top Executives and Technology Investors Top Executives refer to those with biotechnology background, and has experience in serving as officer of chief executive (CEO) or manager; Technology Investors refer to those acquire shares through exchange of technology. As biopharmaceutical industry possesses a unique business model that demands intensive technology, whether top executives and technology investors are willing to participate in a high risk business and satisfy the needs of industry becomes a critical issue. Consequently, Article 7 of the Act stated that 「In order to encourage top executives and technology investors to participate in the operation of Biotech and New Pharmaceuticals Companies and R&D activities, and to share their achievements, new shares issued by a Biotech and New Pharmaceuticals Company to top executives and technology investors (in return of their knowledge and technology) shall be excluded from the amount of their consolidated income or corporate income of the then current year for taxation purposes; provided, however, that if the title to the aforesaid shares is transferred with or without consideration, or distributed as estate, the total purchase price or the market value of the shares at the time of transfer as a gift or distribution as estate shall be deemed income generated in that tax year and such income less the acquisition cost shall be reported in the relevant income tax return.」 Additionally, 「For the title transfer of shares under the preceding paragraph, the Biotech and New Pharmaceuticals Company concerned shall file a report with the local tax authorities within thirty 30 days from the following day of the title transfer.」 Purpose of this regulation is to attract top executives and technology personnel for the company in long-term through defer taxation. Moreover, the Biotech and New Pharmaceutical Company usually caught in a prolong period of losses, and has trouble financing through issuing new shares, as stipulated par value of each share cannot be less than NTD $10.Thus, in order to offer top executive and technology investors incentives and benefits under such circumstances, Article 8 has further provided that」Biotech and New Pharmaceutical Companies may issue subscription warrants to its top executives and technology investors, provided that the proposal for the issuance of the aforesaid subscription warrants shall pass resolution adopted by a majority votes of directors attended by at least two-thirds (2/3) of all the directors of the company; and be approved by the competent authorities. Holders of the subscription warrants may subscribe a specific number of shares at the stipulated price. The amount of stipulated price shall not be subject to the minimum requirement, i.e. par value of the shares, as prescribed under Article 140 of the Company Act. Subscription of the shares by exercising the subscription warrant shall be subject to income tax in accordance with Article 7 hereof. if a Biotech and New Pharmaceutical Company issue new shares pursuant to Article 7 hereof, Article 267 of the Company Act shall not apply. The top executives and technology investors shall not transfer the subscription warrant acquired to pursuant to this Article.」 These three types of tax benefits are detailed incentive measures tailor to the biopharmaceutical industry. However, what is noteworthy is the start date of the benefits provided in the Act. Different from the Statue, the Act allows company to enjoy these benefits when it begins to generate profits, while the Statute provides company tax benefits once the authority approved its application in the current year. Thus, Biotech and New Pharmaceuticals Company enjoys tax benefits as the company starts to make profit. Such approach reflects the actual business operation of the industry, and resolves the issue of tax benefits provided in the Statue is inapplicable to the biopharmaceutical industry. 3 、 Technical Assistance and Capital Investment Due to the R&D capacity and research personnel largely remains in the academic circle, in order to encourage these researchers to convert R&D efforts into commercial practice, the government intends to enhance the collaboration among industrial players, public institutions, and the research and academic sectors, to bolster the development of Biotech and New Pharmaceuticals Company. However, Article 13 of Civil Servants Service Act prohibits officials from engaging in business operation, the Act lifts the restriction on civil servants. According to Article 10 of the Act provided that」For a newly established Biotech and New Pharmaceuticals Company, if the person providing a major technology is a research member of the government research organization, such person may, with the consent of the government research organization, acquired 10% or more of the shares in the Biotech and New Pharmaceuticals Company at the time of its establishment, and act as founder, director, or technical adviser thereof. In such case, Article 13 of the Civil Servants Service Act shall not apply. And the research organization and research member referred to thereof shall be defined and identified by the Executive Yuan, in consultation with the Examination Yuan.」 This regulation was enacted because of the Civil Servants Services Act provided that public officials are not allowed to be corporate shareholders. However, under certain regulations, civil servants are allowed to be corporate shareholders in the sector of agriculture, mining, transportation or publication, as value of the shares cannot exceed 10% of the total value of the company, and the civil servant does not served in the institution. In Taiwan, official and unofficial research institution encompasses most of the biotechnology R&D capacity and research personnel. If a researcher is working for a government research institution, he would be qualified as a public servant and shall be governed by the Civil Servants Service Act. As a result of such restriction, the Act has lifted the restriction and encouraged these researchers to infuse new technologies into the industry. At last, for advancing the development of the industry, Article 11 also provided that 」R&D personnel of the academic and research sectors may, subject to the consent of their employers, served as advisors or consultants for a Biotech and New Pharmaceuticals Company.」 4 、 Other Regulations For introducing and transferring advanced technology in support of the biopharmaceutical industry, Article 9 stated that 「Organization formed with government funds to provide technical assistance shall provide appropriate technical assistance as may be necessary.」 Besides technical assistance, government streamlines the review process taken by various regulatory authorities, in order to achieve an improved product launch process result in faster time-to-market and time-to profit. As Article 12 provided that 「the review and approval of field test, clinical trials, product registration, and others, the central competent authorities shall establish an open and transparent procedure that unifies the review system.」

Most read
[　X Close　]
Copyright © 2016 STLI,III. All Rights Reserved.
TOP