AI will keep Japan at the forefront of digital health solutions.
Japanese healthcare is one of the most advanced systems globally. Residents enjoy near 100% (98.3%) universal healthcare coverage and this provides clear results: Japan records one of the longest average life expectancies in the world.17 18 Meanwhile, Japan’s advanced telehealth solutions have contributed to the growth of digital medicine, particularly during the 2020 pandemic.19
The healthcare system is integrating AI into their everyday functioning. Researchers at the Cancer Institute Hospital in Tokyo have developed AI models for gastric cancer detection that outperform human endoscopists, achieving up to 98.6% sensitivity for large tumours.20 In Kyoto University Hospital, AI is summarising doctors’ notes and discharge summaries with a 92% accuracy rate, significantly reducing paperwork for medical professionals.21
The public also support AI use in triage and diagnosis, especially when combined with human reviews.
of people say they would support the greater use of initial AI triage and diagnosis. This increases to
who would support AI in this case if it was reviewed by a human doctor.
Google’s Project VOICE is using AI to support inclusive communication.
Project VOICE (Valuing Our Individual Communication Expression) is a Google research initiative aimed at improving communication for people with speech and motor impairments, such as ALS (Amyotrophic Lateral Sclerosis) or SMA (Spinal Muscular Atrophy), by helping them express themselves more easily and naturally with the power of AI.22
Built on Google’s Gemini model, Project VOICE uses predictive text technology to suggest likely next words or phrases based on the user’s input and context. By allowing users to select from these suggested options, it enables messages to be conveyed with significantly fewer steps than before, supporting smoother, real-time conversations. The system adapts to each user’s interests and communication style, offering personalised suggestions. It also supports a range of input methods, including eye-tracking and switch controls, to meet different accessibility needs.
While still evolving, Project VOICE is already making it easier for users to express themselves and stay connected to the people around them, and is aiming to set a new standard for inclusive communication technologies. In June 2025, Project VOICE was open-sourced allowing developers worldwide to freely integrate the system into their own products and services.
AI will speed up drug discovery.
Japan remains highly reliant on cutting-edge drugs developed abroad. One of the key drivers of this reliance has been a historical “drug lag” – the time between a drug’s global approval and its availability in Japan. When new therapies take significantly longer to gain domestic approval, Japanese patients and providers often turn to imported treatments already available in global markets.23 This has contributed to a persistent trade deficit in pharmaceuticals, with imports exceeding exports by over JPY 3 trillion (USD 19.7 billion).
The country has taken steps to reduce this lag: while the average delay was 4.5 years between 2008–11, it narrowed to 1.8 years by 2016–19.
AI could make domestic drug discovery easier and faster – further reducing the country’s reliance on international solutions. New AI driven simulation tools such as Google DeepMind’s AlphaFold make it possible to digitally simulate how different molecules interact with each other. This can cut the time to discover new lead candidates, in some cases from years to weeks or even days.
By integrating AI tools into pharmaceutical R&D, Japan’s biotech firms could reduce the average time for drug discovery by
Average delay between drug availability on global market and availability in Japan (years).
Source: PMDA
Medical researchers are partnering with Google to accelerate drug discovery.
Chugai Pharmaceutical, one of the world’s leading digital innovators for medical research and development, has partnered with Google to accelerate drug discovery and advance biomedical research. At the heart of this collaboration is AlphaFold, the groundbreaking AI system developed by Google DeepMind that predicts protein structures with remarkable accuracy. Building on this breakthrough, Chugai is developing a scalable, cloud-based version of AlphaFold2 capable of modelling thousands of proteins each day. This will enable scientists to simulate how antibodies interact with disease targets, potentially reducing the typical 10–15-year timeline for developing new drugs.24
Additionally, Chugai is creating a unified cloud infrastructure to support AI-driven research across the organisation. Leveraging Google Cloud services such as BigQuery and Cloud Run, the platform will empower teams to process large datasets, build internal tools, and deploy machine learning models at scale. Together, these initiatives are central to medical researchers’ digital strategy and their long-term vision to deliver faster, more impactful healthcare breakthroughs.
AI can make the public sector even more efficient.
The public sector, while employing just 5% of the Japanese workforce, is responsible for 21% of value added to the economy.25 26 It is among the top 3% most effective governments globally.27
Yet there is still room to further improve in the Japanese public sector efficiency. Despite its strong international rankings, analogue technology is still widely used and may be holding back further productivity gains in the public sector. 80% of Information and Communication Technology (ICT) investment in
Japan goes towards maintaining legacy technology, which several intergovernmental procedures still rely on.28 29
Making the right investments in AI infrastructure could create a more transparent, flexible and responsive public sector, where repetitive tasks are automated, freeing up worker time for higher value activities.
In total, we estimate that AI could help increase productivity in the public sector in Japan by
Google is delivering practical AI training for public sector workers.
Realising the full potential of AI will require parallel investment in human capital. To this end, Google has introduced the “Local Growth Package,” a comprehensive program designed to empower Japan’s public sector.30
This package is built on two pillars, one of which is a suite of targeted human resource development programs. Key among these is the “AI Connect Academy,” which provides practical, hands-on training for public sector professionals such as local government and central ministry staff. It uses real-world case studies to demonstrate how AI tools can be applied to municipal services and policy challenges.
Complementary initiatives, including the “Gemini Academy” for educators and specialised AI and cybersecurity training for SMEs and startups, are designed to cultivate a robust and skilled talent base across the nation.
AI can help Japanese creators go global.
Generative AI translation tools could help creators from Japan reach another
people globally.
Share of Global Creative Economy by Country.
Source: METI
AI will reduce the costs of cyber fraud.
Japan has been experiencing a surge in cyber fraud in recent years. As of February 2024, an average of more than 1,000 weekly cyberattacks were recorded. This includes financial fraud, information theft, and ransomware attacks. In 2023, one of Japan’s most critical cyber attacks breached Tokyo’s defense networks, revealing major gaps in Japan’s attack preparedness policies.34 The National Cybersecurity Index ranks Japan as having weaker preparedness than countries like Korea, India, and Singapore.35
AI-powered tools can proactively monitor and flag emerging digital security threats. AI can analyse global threat intelligence data to identify trends and potential risks before they become active threats. It can also act as a safety net for phishing and similar attacks and automate threat responses, reducing any potential damage.
Once widely deployed, we estimate that AI could prevent
of the costs from cybersecurity threats and fraud.
Global cyber security rank (out of 176 countries).
A lower rank = poorer preparedness (i.e. 176 = lowest rank).
Source: National Cybersecurity Index