EN, essay

AI Governance in 2019: A Year in Review in Japan

There are two characteristics of Japanese AI Governance in 2019. First, the industries/sectors began to take the lead on the AI governance framework; second, machine learning and fairness became more important.


Since 2016, the government took the lead in discussions about AI governance and ethics in Japan [1]. The Institute for Information and Communications Policy (IICP) the Ministry of Internal Affairs and Communications (MIC), since 2016, has held the “Conference toward AI Network Society”—a conference of advisory experts studying social, economic, ethical, and legal issues towards promoting AI networking in society. The conference released the “AI R&D Guidelines” in 2017 [2] and “AI Utilization Guidelines” in 2019 [3]. Additionally, the Ministry of Economy, Trade and Industry (METI) released the Contract Guidelines on Utilization of AI and Data in 2018 for the development and use of AI-technology software [4]. Culminating from inter-governmental and multi-stakeholder discussions, the “Social Principles of Human-Centric AI” was released from the Cabinet Secretariat in February 2019 [5] and, as a G20 hosting country in June 2019, Japan introduced the principles. At the G20, the AI Principles were endorsed by the G20 Digital Economy Ministers [6].


The “Social Principles of Human-Centric AI” outlines AI governance, allowing industries and sectors to turn its principles into practice. For example, the Japan Business Federation (Keidanren) released the “AI Utilization Strategy: For an AI-Ready Society” that developed an AI use strategy framework in February 2019 [7]. The term “AI-ready Society” is also used in the “Social Principles of Human-Centric AI,” which demonstrates the ties between the government and industries. Companies such as Fujitsu [8], NEC [9], and NTT Data [10] also released AI principles in spring 2019. Both traditional companies and a startup company (ABEJA) organized ethics committees to begin discussions on AI governance and ethics [11].

Two Social Incidents in Japan

While industries commenced the discussion, two incidents in 2019 caught the public’s attention and accelerated the importance of discussing AI governance and its social implications.

AI profiling service by a recruitment managing company

First, there was a scandal involving a recruitment management company selling users’/students’ data to client companies in August [12]. The company developed a profiling service that collects and analyzes job seeker data and calculates the probability score of users declining informal job offers from a specific company. Although the main problem was related to the illegality of using personal information and not the algorithmic bias of AI, this incident was almost the first case in the media involving ethical and legal issues around AI in Japan.

Hate speech by an AI researcher

The second incident occurred in November, where the Project Associate Professor at the University of Tokyo (a director of an AI company) tweeted racist opinions regarding the company’s recruitment policy, and claimed his discriminatory comments were caused by machine learning. The University of Tokyo immediately released its official statement that his tweets contravene the ideals of the University of Tokyo Charter [13]. However, this incident engendered mistrust of the AI research community and questioned how to deal with the issue of bias made by AI, especially machine learning technology.


These incidents raised social anxieties towards machine learning and the responsibilities of research communities. In response, three academic communities that were engaged in machine learning released the “Statement on Machine Learning and Fairness” in December, declaring that (1) machine learning is nothing more than a tool to assist human decision making, and (2) machine learning researchers are committed to improving fairness in society by studying the possible uses of machine learning. This research group will organize a symposium in January 2020 to open a dialogue on machine learning and fairness, involving Machine Learning research communities and other organizations such as social science societies, business associations, and international organizations [14].

Reviewing 2019 and Foreseeing Next Year

Regarding AI governance and ethics, 2019 in Japan has shown that the lead role in these factors has shifted from the government to business. Simultaneously, the social implementation of AI progresses and, consequently, the ethical, legal, and social concerns regarding AI and machine learning have emerged in Japan. However, multi-stakeholder and inter-disciplinary networks on AI governance and ethics have been organized in Japan since 2016, and we will continue to tackle these issues and contribute to the world’s AI governance and ethics discussions.


  1. Arisa Ema, EADv2 Regional Reports on A/IS Ethics: Japan, 2017, https://standards.ieee.org/content/dam/ieee-standards/standards/web/documents/other/eadv2_regional_report.pdf
  2. The Conference toward AI Network Society, Draft AI R&D Guidelines for International Discussions, 2017, https://www.soumu.go.jp/main_content/000507517.pdf
  3. The Conference toward AI Network Society, AI Utilization Guidelines, 2019, https://www.soumu.go.jp/main_content/000658284.pdf
  4. METI, English Translation Version of the Contract Guidelines on Utilization of AI and Data Released, 2019, https://www.meti.go.jp/english/press/2019/0404_001.html
  5. Cabinet Secretariat, Social Principles of Human-Centric AI, 2019, https://www.cas.go.jp/jp/seisaku/jinkouchinou/pdf/humancentricai.pdf
  6. G20 Ministerial Statement on Trade and Digital Economy, 2019, https://g20-digital.go.jp/asset/pdf/g20_2019_japan_digital_statement.pdf
  7. Keidanren, AI Utilization Strategy: For an AI-Ready Society, 2019, http://www.keidanren.or.jp/en/policy/2019/013_outline.pdf
  8. Fujitsu, Fujitsu Formulates “Fujitsu Group AI Commitment,” Demonstrating Dedication to Safe and Secure Use of AI, 2019, https://www.fujitsu.com/global/about/resources/news/press-releases/2019/0313-01.html
  9. NEC, NEC Unveils “NEC Group AI and Human Rights Principles,” 2019, https://www.nec.com/en/press/201904/global_20190402_01.html
  10. NTT DATA, NTT DATA Introduces AI Guidelines, 2019, https://www.nttdata.com/global/en/media/press-release/2019/may/ntt-data-introduces-ai-guidelines
  11. ABEJA, Inauguration of “Ethical Approach to AI”, a committee with a blend of intellectuals to discuss ethical, legal and societal issues in AI, 2019, https://abejainc.com/en/news/article/20190725-411
  12. The Japan Times, Recruit Career urged to make amends for selling job-seekers’ browsing data without consent, 2019, https://www.japantimes.co.jp/news/2019/08/27/business/corporate-business/recruit-career-job-data-scandal/#.XgjHokf7Sbi
  13. The University of Tokyo III/GSII, A message for students about inappropriate writings by a part-time project faculty of III/GSII, 2019, http://www.iii.u-tokyo.ac.jp/news/2019112811043
  14. Symposium and Statement on Machine Learning and Fairness, 2019, http://ai-elsi.org/archives/980

Written by Arisa Ema