Governing AI in Society 5.0
Hideaki Shiroyama, professor of public administration at the University of Tokyo, details Japan’s domestic challenges
and international strategy as it works to implement its vision for the society of the future
Emerging technologies such as artificial intelligence (AI) can be expected to spur innovation that leads to the recovery from economic stagnation and facilitates social inclusion. Society 5.0 is a social vision espoused in various policy documents that refer to emerging technologies such as AI. Those policy documents focus mainly on the use of integrated technologies of physical space and cyberspace, such as the Internet of Things, in tackling social problems.
In implementing emerging technologies in the real world, it is important to secure the trust of the public by setting guidelines and establishing principles. In response to the need to propose AI research and development principles at the meeting of G7 ministers responsible for information and communications technologies in Takamatsu in April 2016, Japan’s Ministry of Internal Affairs and Communication set up the Conference on Networking among AIs in February 2016, which proposed a tentative draft. In October 2016, the Conference on Networking among AIs was reorganised into the Conference Towards an AI Network Society, which produced a further draft set of AI research and development guidelines in July 2017.
They established the principles of collaboration, transparency, controllability, safety, security, privacy, ethics, user assistance and accountability. The conference also introduced draft AI utilisation principles in July 2018.
The discussion about the governance of AI in Japan seems to occupy an intermediate position between those of the United States and the European Union. In the United States, the Executive Order on Maintaining American Leadership in Artificial Intelligence was announced in February 2019. It emphasises driving technological breakthroughs and developing technical standards, although it also mentions fostering public trust and confidence. Policy discussion about AI governance in Japan, by contrast, pays more attention to the need for trust and confidence.
In the European Union, the General Data Protection Regulation came into effect in May 2018, and the Ethics Guidelines for Trustworthy AI were released in April 2019. The guidelines set out seven requirements (see below). The current focus in the policy discussion in Japan, however, is on non-regulatory measures rather than on regulatory measures as seen in Europe. However, the requirements of the European Union’s Ethics Guidelines are similar to the utilisation principles promoted by Japan’s Ministry of Internal Affairs and Communication.
There remain several domestic challenges in developing AI policy in Japan. First, there is the issue of relations among different government ministries. The cabinet office and Ministry of Internal Affairs and Communication have played important roles in setting general principles for developing AI policy. In addition, the Ministry of Economy, Trade and Industry is active in policy for competition and model contract areas. Sectoral ministries such as the Ministry of Land, Infrastructure and Transport and the Ministry of Health, Welfare and Labour are active in each application area (such as autonomous vehicles and medical uses for AI).
Second, the discussions are entering a phase of practice in concrete applications. There is an emerging need to cope with different contexts (including diverse stakeholders) in different sectors. This shift also affects the relations among these ministries.
In parallel with the domestic discussion on AI governance, Japan has attempted to facilitate discussion in the multilayered multilateral arena, utilising its intermediate position between the United States and the European Union and its geopolitical position in Asia.
At the G7 meeting it hosted in 2016, Japan took the initiative to put AI on the policy agenda. It also took the R&D guideline proposal to the Organisation for Economic Co-operation and Development. In addition, in his address to the World Economic Forum in Davos in January 2019, Japanese prime minister Shinzo Abe called for data governance, which he calls ‘data free flow with trust’, under the umbrella of a reformed World Trade Organization.
There maybe some debate about whether the WTO, especially its forum on e-commerce, is the appropriate arena for discussing the free flow of data, but the WTO is clearly one of the most important arenas for such discussions beyond like-minded countries. Prime Minister Abe is calling for the ‘Osaka Track’ for data governance to begin at the G20 summit in June. Securing the agreement in this diverse environment of other countries, including emerging countries such as China, is a difficult task. But the discussion on the Digital Economy Development and Cooperation Initiative, which focused on inclusion, was already undertaken at the G20’s Hangzhou Summit in China in 2016. It might be possible to extend that discussion at Osaka.
The Ethics Guidelines for Trustworthy AI
Human agency and oversight
Technical robustness and safety
Privacy and data governance
Transparency
Diversity, non-discrimination and fairness
Environmental and social well-being
Accountability
