Japan is the first country to tackle many social issues confronting a mature society, such as the declining birthrate and aging population, labor shortage, rural depopulation, and increased fiscal spending. Artificial Intelligence (AI) is considered a key technology to rescue society from these problems. (Government of Japan Citation2019: 1)
This quote, taken from the “Social Principles of Human-Centric AI,” a key Japanese policy document on the ethics of AI, captures something of the growing spectre of demographic crisis haunting the formulation of government policy. Yet it also encapsulates soaring expectations around AI and related technologies, including robotics and the Internet of Things (IoT), that have developed over the past two decades—expectations that have been turbocharged by the arrival of generative AI. The Japanese government is not alone in linking impending calamity with the presumed salvational potential of data-driven technologies. As South Korea’s President Moon put it in the introduction to his government’s 2019 National Strategy for AI:
AI is the culmination of a human dream to constantly complement shortcomings and become more perfect … AI will not only affect industrial sectors but also solve many issues facing our society: public health in an aging society [and] welfare for senior citizens living alone. (Government of Korea Citation2019: 4)
Both the Korean and Japanese governments routinely employ techno-utopian rhetoric in describing these technologies, portraying halcyon visions of positive social and economic transformation, while critical voices tend to be less prominent compared with the United States, Europe, and the United Kingdom.Footnote1
This optimistic rhetoric, inflected with just a hint of desperation, has been backed up by policy action. Both the South Korean and Japanese governments are heavily invested, both financially and politically, in the idea of digital technologies including robotics and AI “rescuing” their societies by solving the “problems” they associate with aging populations, including rising health and care costs, a shrinking workforce, low productivity, rural depopulation, and falling tax revenues. At the same time, both are also vying for lead positions in the development of technical standards, global research and development, and international trade in these sectors, amid growing competition between the United States, the European Union, and China. Even while developing sets of high-level ethical principles for AI, both Japan and South Korea have strenuously pushed back against any immediate attempts at regulation. As Moon put it, the governance of AI should be “defined by the principle of ‘approve first, regulate later’” (Government of Korea Citation2019: 4).Footnote2
In Japan, government policy took a decisive robotic turn under Prime Minister Shinzo Abe. In 2007, during Abe’s first term, his government published “Innovation 25,” a strategy document that presented a highly technologized future vision of Japanese society including the use of robots in healthcare (Robertson Citation2018). National projects were commissioned to lay the regulatory and standards foundations for the expected large-scale implementation of care robots, for example via the 2009–13 US$60 million “Project for the practical utilisation of personal care robots” and the 2013–17 US$125 million “Project for the development and promotion of the introduction of robot care devices” (Wright Citation2023). Abe announced the establishment of a “Robot Revolution Initiative” in 2015, promising to “spread the use of robotics from large-scale factories to every corner of our economy and society” (cited in Bremner Citation2015).
The following year, the Cabinet Office’s 5th Science and Technology Basic Plan, which set out the Japanese central government’s agenda for science and technology for the next five years, introduced the concept of Society 5.0. This was described as “an ideal form of our future society,” a “world-leading super smart society” (Government of Japan Citation2015), and “[a] human-centered society that balances economic advancement with the resolution of social problems by a system that highly integrates cyberspace and physical space.”Footnote3 Society 5.0 would be built by combining AI, big data, robotics, and IoT, increasingly grouped together under the label of “dx” (digital transformation), and this remains a key aspiration in Japanese science, technology and innovation policy.
The healthcare and long-term care sectors were presented as priority areas for the development and governance of AI in the 2017 AI Technology Strategy and Industrialization Roadmap. Over the following years, a series of policy documents set out ethical guidelines for developers and users of AI systems, and this preparatory effort culminated in the release of Japan’s National AI Strategy in 2019—the same year as South Korea’s. In addition, a newly created Cross-ministerial Strategic Innovation Promotion Program launched its US$100 million “Innovative AI Hospital System” project in 2018 as one of 12 initiatives aimed at establishing Society 5.0 (Nakamura Citation2022).
The government’s US$665 million Moonshot Research and Development Program was launched in the following year and included three highly ambitious goals related to health and care.Footnote4 In 2021, the government established the Digital Agency, with the aim of promoting future-oriented digital transformation and the creation of a digital society. However, several recent high-profile, state-led initiatives such as the development and rollout of the COVID contact tracing app COCOA and attempts to integrate health insurance cards into the digital My Number ID card system have run into difficulties, including major development errors and data breaches that reduced public trust and fed resistance to digitisation (Wright et al. Citation2021)—belying the idea of a public frictionlessly accepting the technological infrastructures required for Society 5.0.
Robust policy action has also been underway in South Korea. Following President Moon Jae-In’s election in 2017, his administration launched a Presidential Committee on the Fourth Industrial Revolution, which focused on a framework of “DNA”: data, networks, and AI. Digital healthcare formed one of the four pillars of this strategy, and has been seen as a key growth area for Korean business. The vast majority of hospitals and clinics in South Korea have adopted electronic patient records, with over 90% using them by 2014 (Park and Han Citation2017).
The South Korean government’s strategy has involved pursuing an open data ecosystem in line with its endorsement of the Open Data Charter in 2015, and loosening the country’s strict regulations on data protection to enable greater data sharing, while making substantial investments in start-ups and research and development projects to further grow the digital healthcare sector. For example, the 2018 government project FEEDER-NET encouraged participating hospitals to share electronic patient records in a common format and develop an open research platform. Plans were also announced to build a database for medical data incorporating the genetic and biometric data of 10 million patients in order to enable digital health companies to develop new treatments and services (Intralink Citation2019).
In 2019, Moon introduced his government’s National Strategy for AI with a story:
At 3:40 am one early morning in May this year, an elderly man called out ‘help me’ to an AI speaker as he collapsed from high blood pressure. The virtual assistant recognized his shout as an emergency signal, alerted a 119 rescue squad and saved his life. (Government of Korea Citation2019: 3)
From this dramatic opening, Moon pledged to transform his government into “an AI-based digital government” (Government of Korea Citation2019: 5), and set out detailed plans and aspirations for AI, including its use in bio and medical sectors, and the implementation of care robots.
In 2020, the Moon administration announced its five-year, US$95 billion plan for a Korean New Deal, which centered on 10 major projects including smart healthcare, AI government, and a “data dam”—a large store of big data to support research and development. This included plans to build 18 “smart hospitals,” develop the ability to diagnose 20 diseases using AI, and provide IoT sensor devices and “AI speakers that engage in conversations to enhance cognitive functions” to 120,000 older people, as well as wearable health monitoring devices to 200,000 people with chronic conditions (Government of Korea Citation2020: 21). These were viewed as part of a broad investment in “the ‘untact’ industry,” where “untact” refers to digital self-service that does not require face-to-face contact with another person.
The South Korean government’s remarkably successful response to the COVID-19 pandemic in 2020 and 2021 was seen by some as a vindication of this data-centric untact approach to service provision in healthcare and beyond (Lee and Lee Citation2020). For roughly similar periods from the start of the pandemic to late 2021 (and early 2022 in the case of the US figures), the US recorded 301 excess deaths per 100,000 people;Footnote5 for Japan the figure was 16, and for South Korea the figure was just 5 (Economist Citation2022). By 2021, the Korean government’s research and development budget for digital health was US$850 million (Lee Citation2021; cited in Jung Citation2021), and in 2023, the “My HealthWay” app was launched to bring together a user’s medical records and consumer healthcare data held by hundreds of medical facilities in a single app interface (Ang Citation2023).
In both Japan and South Korea, much of the impetus for and investment in these seemingly transformative technologies have come from the government, and the results are not yet clear: for example, it is notable that the state’s focus on developing the technologies and market for care robots in both countries has not yet translated into widespread use, as uptake has lagged far behind the socio-technical imaginaries conjured up by policymakers (Wright Citation2023). As one commentator put it in relation to South Korean state investment, “major spending on innovative technologies like 5G and AI is a bet in the face of a looming demographic crisis—a prudent one, but a bet nonetheless” (Ferrier Citation2021).
Yet this bet looks indeed increasingly prudent in light of the meteoric rise of generative AI since the release of ChatGPT in November 2022, which has raised global debates about the promises and perils of AI to a fever pitch. Large language models have demonstrated to hundreds of millions of users their potential to disseminate medical knowledge via accessible chatbot interfaces, personalize medicine and care, and greatly reduce time spent on admin, in addition to emerging specialist uses in medical research and drug development, and diagnosis and medical imaging.
At the same time, they perpetuate or exacerbate existing concerns around the potential for AI to deliver biased, discriminatory, unfair, inaccurate or misleading outputs, concerns around privacy and data security, questions about accountability and liability, particularly as models are often closed and opaque, and about the scale of any harms now that tools like ChatGPT have so many users. This in turn links to concerns about concentrations of power and capital among a small handful of AI companies mainly located in the US and China that are producing the most widely-used foundation models.
As both Japan and South Korea pursue courses aimed at ever more ambitious technological innovation to counterbalance aging and shrinking populations, the extent to which these projects will prove successful in the age of generative AI and what “success” would look like, what new socio-technical formations will be mainstreamed, and what implications digitalisation and roboticisation will have for people’s health and well-being as well as for wider society remain to be seen.
This special issue brings together articles from a range of scholars from science and technology studies, nursing, anthropology, and social policy, to critically examine how AI and related technologies such as socially assistive robots are imagined or expected to transform futures of health and social care in Japan and South Korea, how they are being deployed, and how actors in the government, industry, and civil society propose that they be governed.
Papers by James Wright and Do Hyun Park, Eunjung Cho, and Yong Lim, provide overviews of the landscape of AI ethics and governance in both countries. On the Japanese side, James Wright examines guidelines and policy documents on the ethics and governance of AI released by the Japanese government over the past decade as well as drawing on semi-structured interviews with key architects of Japan’s AI policies. Wright traces the development of AI ethics at the national level, and considers what this process reveals about broader Japanese state imaginaries of how advanced technology should be developed and used, and what a future with these technologies should look like.
On the South Korean side, Do Hyun Park, Eunjung Cho, and Yong Lim explain how AI governance began to emerge as a focal point of discourse in Korea from the mid-2010s, with government and public bodies publishing guidelines and engaging in international efforts to develop ethical recommendations. They describe the “tough balancing act” involved in AI governance, as policymakers have aimed to promote innovation in the industry while ensuring the safety of these technologies.
Back to the Japanese context, Anne Aronsson’s paper addresses the concept of robotics as a media technology, and the role of human agency in shaping imagination as an interpretive framework. Adapting an approach by anthropologist Tom Boellstorff, whose work focuses on the reality of virtual worlds, her paper tackles ontological questions regarding the real of robot care.
Heesun Shin and Chihyung Jeon’s paper offers insights into South Korea’s proposed technical solutions for an ageing population. Their case study is Hyodol, a robot developed for older adults living alone which can provide a variety of services including dementia prevention quizzes and daily medication reminders. As such, it emerges as a “robot grandchild” to which users become emotionally attached. The deployment of Hyodol also serves as the basis for their compelling ethnographic documentary film, Hug Me Tight: A Robot Story, a still image of which is featured on the front cover of this special issue.
Remaining in the context of social care, Giulia De Togni’s paper, based on extensive ethnographic fieldwork in Japan, explores the performance and ethical implications of technoscientific imaginaries portraying socially assistive robots (SARs) as already reliable, safe, and efficient. Her paper sheds light on the intricate relationship between science, technology, the state, and society, emphasizing their use as instruments of power for state-led national development objectives in Japan.
Seonsam Na and Eunjeong Ma offer a detailed analysis of robot-assisted rehabilitation in South Korea. They focus on the challenges posed by critical staff shortages in a rapidly aging society. Their paper reveals that rehabilitation medicine settles uneasily on the notion of neuroplasticity as a theoretical tool to legitimize robot-assisted sessions, in the absence both of upstream treatment options such as stem cell therapy and of their discernible benefits over human-based intervention.
Finally, Yulia Frumer and Selma Šabanović provide a critical commentary on the articles presented in this special issue. They focus in particular on the mise-en-scène nature of robotics and its implications, examining how staging, both literally and metaphorically, shapes the public’s perception of robots’ capabilities and autonomy. While calling attention to the need for critical analysis of the hidden human labor behind robotic (and AI) performances, they highlight how such staging influences imaginings about technology’s role in society, particularly in the context of healthcare robotics in East Asia.
This special issue aims to offer novel insights and innovative multidisciplinary methodological approaches to the study of AI and SARs in the health and social care sectors. The ground covered by these papers will move these topics forward in terms of both area studies (Japan and South Korea being the countries we focused on) and science and technology studies more broadly. Policy momentum for implementing these technologies continues to grow, and the widespread and lightning-fast adoption of tools like ChatGPT demonstrates that a broad swathe of the public is already experimenting with them in everyday use, but despite previous work on areas of healthcare innovation, there remains a lack of synthesis from the social sciences to account for how ethics, governance and practices of care will be influenced and potentially transformed as these technologies enter different care spaces.
To this end, the papers in this special issue critically address the following timely questions: What forms of AI and related technologies, such as robotics, are actually being developed and deployed in health and care? How does the aspirational rhetoric of these technologies connect with realities of use? How are the ethics and governance of these systems being conceptualized, drawn up into guidelines and principles, and operationalized? Where is the “human” in “human-centric” robotics and AI?
As discourses around the future potential and imagined impacts of these technologies are reaching new heights of aspiration and hype globally, not only in Japan and South Korea, these papers bear importance beyond the regionally specific case studies examined here. We offer a rich analysis that can be used to draw a comparison between Japan/South Korea and other countries’ politics of innovation, governance, and ethics. This analysis will further our understanding of the development and application of AI and robotic innovation in the contested area of human care.
