> Article > Science & Technology R&D Law
Back
Previous Taiwan Recent Regulatory Development- Promoting Biotech and New Pharmaceuticals Industry Next Analyzing the Framwork of the Regulation「Act For The Development of Biotech And New Pharmaceuticals Industry」in Taiwan

Introducing and analyzing the Scope and Benefits of the Regulation「Statute for Upgrading Industries」in The Biotechnology Industry in Taiwan

The recent important regulation for supporting the biopharmaceutical industry in Taiwan has been the 「Statute for Upgrading Industries」 (hereinafter referred to as 「the Statute」).The main purpose of the Statue is for upgrading all industry for future economic development, so it applies to various industries, ranging from agriculture, industrial and service businesses. In other words, the Statute does not offer incentive measures to biopharmaceutical industry in particular, but focuses on promoting the industry development in general.

Statute for Upgrading Industry and Related Regulations

Generally speaking, the Statute has a widespread influence on industry development in Taiwan. The incentive measures provided in the Statute is complicated and covered other related regulations under its legal framework. Thus, the article will be taking a multi-facet perspective in discussing the how Statute relates to the biopharmaceutical industry.

1 、 Scope of Application

According to Article 1 of the Statute, the term 「industries」 refers to agricultural, industrial and service businesses. Consequently, nearly all kinds of industries fall under this definition, and the Statute is applicable to all of them. Moreover, in order to promote the development and application of emerging technology as well as cultivating the recognized industry, the Statute provides much more favorable terms to these industries. These emerging and major strategic industries includes computer, communication and consumer electronics (3C), precise mechanics and automation, aerospace, biomedical and chemical production, green technology, material science, nanotechnology, security and other product or service recognized by the Executive Yuan.

2 、 Tax Benefits

The Statute offers several types of tax benefits, so the industry could receive sufficient reward in every way it could, and promote a sound cycle in creating new values through these benefits.

 (1) Benefits for the purchase of automation equipment
The said procured equipment and technology over NTD600, 000 may credit a certain percentage of the investment against the amount of profit-seeking enterprise income tax payable for the then current year. For the purchase of production technology, 5% may be credited. For the purchase of equipment, 7% may be credited. And any investment plan that includes the purchasing of equipment for automation can qualify for a low-interest preferential loan. Besides, for science-based industrial company imported overseas equipment that is not manufacture by local manufactures, from January 1, 2002, the imported equipment shall be exempted from import and business tax. And if the company is a bonded factory, the raw materials to be imported from abroad by it shall also be exempt from import duties and business tax.
 (2) Benefits for R&D expenditure
Expenditure concurred for developing new products, improving production technology, or improving label-providing technology may credit 30%of the investment against the amount of profit-seeking enterprise income tax payable for the then current year. Research expenditures of the current year exceeding the average research expenditure for the past two years, the excess in research expenditure shall be 50% deductible. Instruments and equipments purchased by for exclusive R&D purpose, experimentation, or quality inspection may be accelerated to two years. At last, Biotech and New Pharmaceuticals Company engages in R&D activities, such as Contract research Organization (CRO), may credit 30% of the investment against the amount of profit-seeking enterprise income tax payable.
 (3) Personnel Training
When a company trained staff and registered for business-related course, may credit 30% of the training cost against the amount of profit-seeking enterprise income tax payable for the then current year. Where training expenses for the current year exceeds the two-year average, 50% of the excess portion may be credited.
 (4) Benefit for Newly Emerging Strategic Industries
Corporate shareholders invest in newly emerging strategic industries are entitled to select one of the following tax benefits:

  1. A profit seeking enterprise may credit up to 20% of the price paid for acquisition of such stock against the profit seeking enterprise income tax. An individual may credit up to 10%. As of January and once every year, there will be a 1% reduction of the price paid for acquisition of such stock against the consolidated income tax payable in the then current year.
  2. A company, within two years from the beginning date for payment of the stock price by its shareholders, selects, with the approval of its shareholder meeting, the application of an exemption from profit-seeking enterprise income tax and waives the shareholders investment credit against payable income tax as mentioned above. However, that once the selection is made, no changes shall be allowed.

 (5) Benefits for Investment in Equipment or Technology Used for Pollution Control
To prevent our environment from further pollution, the Government offers tax benefits to reward companies in making improvements. Investment in equipment or technology used for pollution control may credit 7% of the equipment expenditure, and 5% of the expenditure on technology against the amount of profit-seeking enterprise income tax payable for the then current year. For any equipment that has been verified in use and specialized in air pollution control, noise pollution control, vibration control, water pollution control, environmental surveillance and waste disposal, shall be exempt from import duties and business tax. And for investment plans that planned implementation of energy saving systems can apply for a low interest loan.
 (6) Incentive for Operation Headquarter
To encourage companies to utilize worldwide resources and set up international operation network, if they established operation headquarters within the territory of the Republic of China reaching a specific size and bringing about significant economic benefit, their following incomes shall be exempted from profit-seeking enterprise income tax:

  1. The income derived from provision of management services or R&D services.
  2. The royalty payment received under its investments to its affiliates abroad.
  3. The investment return and asset disposal received under its investment to its affiliates abroad.

 (7) Exchange of Technology for Stock Option
The emerging-industrycompany recognized by government, upon adoption of a resolution by a majority voting of the directors present at a meeting of its board of directors attended by two-thirds of the directors of the company, may issue stock options to corporation or individual in exchange for authorization or transfer of patent and technologies.
 (8) Deferral of Taxes on the Exchange of Technology for Shares
Taxes on income earned by investors from the acquisition of shares in emerging-industry companies in exchange for technology will be deferred for five years, on condition that the shares exchanged for technology amount to more than 20% of the company's total stock equity and that the number of persons who obtain shares in exchange for technology does not exceed five.

3 、 Technical Assistance and Capital Investment

The rapid industry development has been closely tied to the infusion of funds. In addition to tax benefits, the Statute incorporates regulations especially for technical assistance and capital investment as below:
 (1) In order to introduce or transfer advanced technologies, technical organization formed with the contribution of government shall provide appropriate technical assistance as required.
 (2) In order to advance technologies, enhance R&D activities and further upgrade industries, the relevant central government authorities in charge of end enterprises may promote the implementation of industrial and technological projects by providing subsidies to such R&D projects.
 (3) In order to assist the start-up of domestic small-medium technological enterprises and the overall upgrading of the entire industries, guidance and assistance shall be provided for the development of venture capital enterprises.

※Introducing and analyzing the Scope and Benefits of the Regulation「Statute for Upgrading Industries」in The Biotechnology Industry in Taiwan,STLI, https://stli.iii.org.tw/en/article-detail.aspx?no=105&tp=2&i=168&d=6135 (Date:2025/11/07)
Quote this paper
Release： 2013/04
Tag
Read with phone
Back
Previous Taiwan Recent Regulatory Development- Promoting Biotech and New Pharmaceuticals Industry Next Analyzing the Framwork of the Regulation「Act For The Development of Biotech And New Pharmaceuticals Industry」in Taiwan
You may be interested
Introduction to Taiwan’s Guidelines for Implementing Decentralized Elements in Medicinal Product Clinical Trials

Introduction 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).

Taiwan Planed Major Promoting Program for Biotechnology and Pharmaceutical Industry

Taiwan Government Lauched the “Biotechnology Action Plan” The Taiwan Government has planned to boost the support and develop local industries across the following six major sectors: biotechnology, tourism, health care, green energy, innovative culture and post-modern agriculture. As the biotechnology industry has reached its maturity by the promulgation of "Biotech and New Pharmaceutical Development Act" in July, 2007, it will be the first to take the lead among the above sectors. Thus, the Executive Yuan has launched the Biotechnology Action Plan as the first project in building the leading industry sectors, to upgrade local industries and stimulate future economic growth. Taiwan Government Planed to Promote the Biotechnology and Other newly Industries by Investing Two Hundred Billion To expand every industrial scale, enhance industrial value, increase the value around the main industrial field, and to encourage the industrial development by government investments for creating the civil working opportunities to reach the goal of continuous economic development, the Executive Yuan Economic Establishment commission has expressed that, the government has selected six newly industrials including "Biotechnology", "Green Energy", "Refined Agriculture", "Tourism", "Medicare", and "Culture Originality" on November 19, 2009 to promote our national economic growth. The government will invest two hundred billion NT dollars to support the industrial development aggressively and to enhance the social investments from year 2009 to 2012. According to a Chung-Hua Institution for Economic Research report, the future growth rate will reach 8.16% after the evaluation, Hence, the future of the industries seems to be quite bright. Currently, the government plans to put money into six newly industries through the existing ways for investment. For instance, firstly, in accordance with the "Act For The Development Of Biotech And New Pharmaceuticals Industry" article 5 provision 1 ",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 corporate income tax, enjoy a reduction in its corporate income tax payable for up to thirty-five percent (35%) of the total funds invested in research and development ("R&D") and personnel training each year; provided, however, 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 previous two years, fifty percent (50%) of the amount in excess of the average may be used to credit against the amount of corporate income tax payable. Secondly, according to same act of the article 6 provision 1 ", in order to encourage the establishment or expansion of Bio tech and New Pharmaceuticals Companies, a profit-seeking enterprise that (i) subscribes for the stock issued by a Biotech and New Pharmaceuticals Company at the time of the latter's establishment or subsequent expansion; and (ii) has been a registered shareholder of the Biotech and New Pharmaceuticals Company for a period of three (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 corporate income tax payable for up to twenty percent (20%) of the total amount of price paid for the subscription of shares in such Biotech and New Pharmaceuticals Company; provided that such Biotech and New Pharmaceuticals Company has not applied for exemption from corporate income tax or shareholders investment credit based on the subscription price under other applicable laws and regulations. Thirdly, to promote the entire biotechnological industry development, the government has drafted the "Biotechnology Takeoff Package" for subsidizing the startup´s social investment companies which can satisfy the conditions to invest in "Drug discovery", "Medical Device" or other related biotech industries up to 5 billion with the capital invest in domestic industry over 50%, , with operating experience of multinational biotech investment companies with capital over 150 million in related industrial fields, and with the working experiences of doctor accumulated up to 60 years. Additionally, the refined agriculture industry field has not only discovered the gene selected products, but also combined the tourism with farming business for new business model creation. According to the "Guidelines for Preferential Loans for the Upgrading of Tourism Enterprises" point 4 provision 1, the expenditure for spending on machine, instruments, land or repairing can be granted a preferential loan in accordance with the rule of point 6, and government will provide a subsidy of interest for loaning Tourism Enterprises with timely payments. At last, Council for Economic Planning and Development also points out because most of technology industry has been impacted seriously by fluctuation of international prosperity due to conducting the export trade oriented strategy. Furthermore, the aspects of our export trade of technology industry have been impacted by the U.S. financial crisis and the economic decay in EU and US; and the industrial development seems to face the problem caused by over centralization in Taiwan. Hence, the current framework of domestic industry should be rearranged and to make it better by promoting the developmental project of six newly industries. Taiwan Government Had Modifies Rules to Accelerate NDA Process and Facilitate Development of Clinical Studies in Taiwan In July 2007, the "Biotech and New Pharmaceutical Development Act" modified many regulations related to pharmaceutical administration, taxes, and professionals in Taiwan. In addition, in order to facilitate the development of the biotechnology and pharmaceutical industries, the government has attempted to create a friendly environment for research and development by setting up appropriate regulations and application systems. These measures show that the Taiwanese government is keenly aware that these industries have huge potential value. To operate in coordination with the above Act and to better deal with the increasing productivity of pharmaceutical R&D programs in Taiwan, the Executive Yuan simplified the New Drug Application (NDA) process to facilitate the submission that required Certificate of Pharmaceutical Product (CPP) for drugs with new ingredients. The current NDA process requires sponsors to submit documentation as specified by one of the following four options: (1) three CPPs from three of "ten medically-advanced countries," including Germany, the U.S., England, France, Japan, Switzerland, Canada, Australia, Belgium, and Sweden; (2) one CPP from the U.S., Japan, Canada, Australia, or England and one CPP from Germany, France, Switzerland, Sweden, or Belgium; (3) a Free Sale Certificate (FSC) from one of ten medically-advanced countries where the pharmaceuticals are originally produced and one CPP from one of the other nine countries; or (4) a CPP from the European Medicines Agency. Thus, the current NDA process requires sponsors to spend inordinate amounts of time and incur significant costs to acquire two or three FSCs or CPPs from ten medically-advanced countries in order to submit an NDA in Taiwan. According to the new rules, sponsors will not have to submit above CPPs if (1) Phase I clinical studies have been conducted in Taiwan, and Phase III Pivotal Trial clinical studies have been simultaneously conducted both in Taiwan and in another country or (2) Phase II and Phase III Pivotal Trial clinical studies have been simultaneously conducted both in Taiwan and in another country. Besides, the required minimum numbers of patients were evaluated during each above phase. Therefore, sponsors who conduct clinical studies in Taiwan and in another country simultaneously could reduce their costs and shorten the NDA process in Taiwan. The new rules aim to encourage international pharmaceutical companies to conduct clinical studies in Taiwan or to conduct such studies cooperatively with Taiwanese pharmaceutical companies. Such interactions will allow Taiwanese pharmaceutical companies to participate in development and implementation of international clinical studies in addition to benefiting from the shortened NDA process. Therefore, the R&D abilities and the internationalization of the Taiwanese pharmaceutical industry will be improved.

The Key Elements for Data Intermediaries to Deliver Their Promise

The Key Elements for Data Intermediaries to Deliver Their Promise 2022/12/13 　　As human history enters the era of data economy, data has become the new oil. It feeds artificial intelligence algorithms that are disrupting how advertising, healthcare, transportation, insurance, and many other industries work. The excitement of having data as a key production input lies in the fact that it is a non-rivalrous good that does not diminish by consumption.[1] However, the fact that people are reluctant in sharing data due to privacy and trade secrets considerations has been preventing countries to realize the full value of data. [2] 　　To release more data, policymakers and researchers have been exploring ways to overcome the trust dilemma. Of all the discussions, data intermediaries have become a major solution that governments are turning to. This article gives an overview of relevant policy developments concerning data intermediaries and a preliminary analysis of the key elements that policymakers should consider for data intermediaries to function well. I. Policy and Legal developments concerning data intermediaries 　　In order to unlock data’s full value, many countries have started to focus on data intermediaries. For example, in 2021, the UK’s Department for Digital, Culture, Media and Sport (DCMS) commissioned the Centre for Data Ethics and Innovation (CDEI) to publish a report on data intermediaries[3] , in response to the 2020 National Data Strategy.[4] In 2020, the European Commission published its draft Data Governance Act (DGA)[5] , which aims to build up trust in data intermediaries and data altruism organizations, in response to the 2020 European Strategy for Data.[6] The act was adopted and approved in mid-2022 by the Parliament and Council; and will apply from 24 September 2023.[7] The Japanese government has also promoted the establishment of data intermediaries since 2019, publishing guidance to establish regulations on data trust and data banks.[8] II. Key considerations for designing effective data intermediary policy 1.Evaluate which type of data intermediary works best in the targeted country 　　From CDEI’s report on data intermediaries and the confusion in DGA’s various versions of data intermediary’s definition, one could tell that there are many forms of data intermediaries. In fact, there are at least eight types of data intermediaries, including personal information management systems (PIMS), data custodians, data exchanges, industrial data platforms, data collaboratives, trusted third parties, data cooperatives, and data trusts.[9] Each type of data intermediary was designed to combat data-sharing issues in specific countries, cultures, and scenarios. Hence, policymakers need to evaluate which type of data intermediary is more suitable for their society and market culture, before investing more resources to promote them. 　　For example, data trust came from the concept of trust—a trustee managing a trustor’s property rights on behalf of his interest. This practice emerged in the middle ages in England and has since developed into case law.[10] Thus, the idea of data trust is easily understood and trusted by the British people and companies. As a result, British people are more willing to believe that data trusts will manage their data on their behalf in their best interest and share their valuable data, compared to countries without a strong legal history of trusts. With more people sharing their data, trusts would have more bargaining power to negotiate contract terms that are more beneficial to data subjects than what individual data owners could have achieved. However, this model would not necessarily work for other countries without a strong foundation of trust law. 2.Quality signals required to build trust: A government certificate system can help overcome the lemon market problem 　　The basis of trust in data intermediaries depends largely on whether the service provider is really neutral in its actions and does not reuse or sell off other parties’ data in secret. However, without a suitable way to signal their service quality, the market would end up with less high-quality service, as consumers would be reluctant to pay for higher-priced service that is more secure and trustworthy when they have no means to verify the exact quality.[11] This lemon market problem could only be solved by a certificate system established by actors that consumers trust, which in most cases is the government. 　　The EU government clearly grasped this issue as a major obstacle to the encouragement of trust in data intermediaries and thus tackles it with a government register and verification system. According to the Data Government Act, data intermediation services providers who intend to provide services are required to notify the competent authority with information on their legal status, form, ownership structure, relevant subsidiaries, address, public website, contact details, the type of service they intend to provide, the estimated start date of activities…etc. This information would be provided on a website for consumers to review. In addition, they can request the competent authority to confirm their legal compliance status, which would in turn verify them as reliable entities that can use the ‘data intermediation services provider recognised in the Union’ label. 3.Overcoming trust issues with technology that self-enforces privacy: privacy-enhancing technologies (PETs) 　　Even if there are verified data intermediation services available, businesses and consumers might still be reluctant to trust human organizations. A way to boost trust is to adopt technologies that self-enforces privacy. A real-world example is OpenSAFELY, a data intermediary implementing privacy-enhancing technologies (PETs) to provide health data sharing in a secure environment. Through a federated analytics system, researchers are able to conduct research with large volumes of healthcare data, without the ability to observe any data directly. Under such protection, UK NHS is willing to share its data for research purposes. The accuracy and timeliness of such research have provided key insights to inform the UK government in decision-making during the COVID-19 pandemic. 　　With the benefits it can bring, unsurprisingly, PETs-related policies have become quite popular around the globe. In June 2022, Singapore launched its Digital Trust Centre (DTC) for accelerating PETs development and also signed a Memorandum of Understanding with the International Centre of Expertise of Montreal for the Advancement of Artificial Intelligence (CEIMIA) to collaborate on PETs.[12] On September 7th, 2022, the UK Information Commissioners’ Office (ICO) published draft guidance on PETs.[13] Moreover, the U.K. and U.S. governments are collaborating on PETs prize challenges, announcing the first phase winners on November 10th, 2022.[14] We could reasonably predict that more PETs-related policies would emerge in the coming year. Reference: [1] Yan Carrière-Swallow and Vikram Haksar, The Economics of Data, IMFBlog (Sept. 23, 2019), https://blogs.imf.org/2019/09/23/the-economics-of-data/#:~:text=Data%20has%20become%20a%20key,including%20oil%2C%20in%20important%20ways (last visited July 22, 2022). [2] Frontier Economics, Increasing access to data across the economy: Report prepared for the Department for Digital, Culture, Media, and Sport (2021), https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/974532/Frontier-access_to_data_report-26-03-2021.pdf (last visited July 22, 2022). [3] The Centre for Data Ethics and Innovation (CDEI), Unlocking the value of data: Exploring the role of data intermediaries (2021), https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1004925/Data_intermediaries_-_accessible_version.pdf (last visited June 17, 2022). [4] Please refer to the guidelines for the selection of sponsors of the 2022 Social Innovation Summit: https://www.gov.uk/government/publications/uk-national-data-strategy/national-data-strategy(last visited June 17, 2022). [5] Regulation of the European Parliament and of the Council on European data governance and amending Regulation (EU) 2018/1724 (Data Governance Act), 2020/0340 (COD) final (May 4, 2022). [6] Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and The Committee of the Regions— A European strategy for data, COM/2020/66 final (Feb 19, 2020). [7] Proposal for a Regulation on European Data Governance, European Parliament Legislative Train Schedule, https://www.europarl.europa.eu/legislative-train/theme-a-europe-fit-for-the-digital-age/file-data-governance-act(last visited Aug 17, 2022). [8] 周晨蕙，〈日本資訊信託功能認定指引第二版〉，科技法律研究所，https://stli.iii.org.tw/article-detail.aspx?no=67&tp=5&d=8422（最後瀏覽日期︰2022/05/30）。 [9] CDEI, supra note 3. [10] Ada Lovelace Institute, Exploring legal mechanisms for data stewardship (2021), 30~31,https://www.adalovelaceinstitute.org/wp-content/uploads/2021/03/Legal-mechanisms-for-data-stewardship_report_Ada_AI-Council-2.pdf (last visited Aug 17, 2022). [11] George A. Akerlof, The Market for "Lemons": Quality Uncertainty and the Market Mechanism, THE QUARTERLY JOURNAL OF ECONOMICS, 84(3), 488-500 (1970). [12] IMDA, MOU Signing Between IMDA and CEIMIA is a Step Forward in Cross-border Collaboration on Privacy Enhancing Technology (PET) (2022),https://www.imda.gov.sg/-/media/Imda/Files/News-and-Events/Media-Room/Media-Releases/2022/06/MOU-bet-IMDA-and-CEIMIA---ATxSG-1-Jun-2022.pdf (last visited Nov. 28, 2022). [13] ICO publishes guidance on privacy enhancing technologies, ICO, https://ico.org.uk/about-the-ico/media-centre/news-and-blogs/2022/09/ico-publishes-guidance-on-privacy-enhancing-technologies/ (last visited Nov. 27, 2022). [14] U.K. and U.S. governments collaborate on prize challenges to accelerate development and adoption of privacy-enhancing technologies, GOV.UK, https://www.gov.uk/government/news/uk-and-us-governments-collaborate-on-prize-challenges-to-accelerate-development-and-adoption-of-privacy-enhancing-technologies (last visited Nov. 28, 2022); Winners Announced in First Phase of UK-US Privacy-Enhancing Technologies Prize Challenges, NIST, https://www.nist.gov/news-events/news/2022/11/winners-announced-first-phase-uk-us-privacy-enhancing-technologies-prize (last visited Nov. 28, 2022).

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).

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