This is Hamamoto from TIMEWELL
This is Hamamoto from TIMEWELL.
Looking for AI training and consulting?
Learn about WARP training programs and consulting services in our materials.
AI White Paper: Japan's National Strategy in the New Era of AI (Summary)
AI White Paper: Japan's National Strategy in the New Era of AI (Summary)
The AI White Paper - Japan's National Strategy in the New Era of AI - (Summary Version)
The LDP's "Project Team on the Evolution and Implementation of AI" published a draft "AI White Paper" on March 30.
The AI strategy white paper emphasized the need for a new national strategy. In particular, concerns are growing about the impact of AI on foundational models, and strategies to address this are needed. There were also opinions that more appropriate responses are needed regarding regulations on social acceptance abroad.
It was noted that it is important to strengthen the capabilities needed for AI development and cultivate capacity for AI development. It was also proposed that the utilization of public and private data should be actively promoted.
Regarding the use of AI in both the public and private sectors, it was proposed that local governments utilize the National Strategic Special Zones system to advance AI-powered smart city development. It was also noted that private companies need to promote proactive approaches such as formulating policies for AI utilization and establishing CDO positions.
Finally, the impact of AI on education was also discussed. It was noted that improving AI literacy is important, and it was proposed that AI literacy be incorporated into the public education curriculum.
Going Forward, We Hope to See
Going forward, we hope to see a national strategy for more appropriate AI use formulated and put into practice.
The need for a new national strategy: As AI advances, Japan's AI technology competitiveness is lower than that of Western countries, making it necessary to formulate a comprehensive national strategy. It is necessary to review policies from the perspectives of research and development, economic structure, social infrastructure, talent development, and national security.
Cultivation of AI technology: Strengthen the development capacity for foundational AI models and promote the collection of data resources and talent development. It is necessary to develop an environment that promotes the use of public and private data through AI.
Legal regulation of AI use: Consider legal regulations regarding risks such as human rights violations, national security, and undue interference with democratic processes, and work with overseas regulations to establish a global AI framework. It is also necessary to revise analog-era regulations to be based on digital principles and to establish guidelines for AI use in the education sector.
Promoting and supporting AI utilization in the private sector
- "Almost all white-collar work is likely to be affected." (Professor Yutaka Matsuo, University of Tokyo)
- "The age is coming where 'machines' will generate infinite content." (Professor Kazumi Ataka, Keio University)
- "Something equivalent to the invention of the engine, semiconductor, and internet is happening at explosive speed." (Hiroaki Kitano, President & CEO, Sony CSL)
(2) Western Countries Are Advancing Regulations for Social Acceptance
(2) Western Countries Are Advancing Regulations for Social Acceptance
There is a risk that "reasonable lies" may be mixed into the output of LLMs (language models).
There is a risk that highly sophisticated fake information will spread.
➝ It is time to reassess the distance from Western regulations.
(3) Developing a new national AI strategy suited to the AI era
Intensifying international competition
Japan ranks 29th out of 63 countries in the digital competitiveness ranking.
The United States has a plan to invest $2.6 billion (approximately ¥340 billion) in AI computing resources.
The UK has announced a plan to invest £900 million (approximately ¥145 billion) in LLM development.
Recommendations ○ Taking into account the rapid evolution and social implementation of foundational AI such as LLMs, it is recommended that the government formulate a comprehensive strategy suited to the new AI era in line with the various recommendations set out in this white paper. This requires rapid development of new policies and reassessment of past efforts.
○ In formulating a new national strategy, it is required to have a competitive advantage over other countries in both content and scale. It is necessary to establish an AI policy command center, expand organizational capacity, and actively incorporate the knowledge of domestic and international experts and private companies. A comprehensive and urgent review of policies from a wide range of perspectives — including research and development, economic structure, social infrastructure, talent development, and national security — must be conducted.
(1) Building and strengthening the capacity to develop AI models including foundational models
Recommendations ○ Actively utilize overseas platforms, accumulate domestic knowledge about foundational model AI, and accelerate applied research and development.
○ Make continuous investments and support in developing domestic foundational technology capabilities toward developing foundational models and other AI models.
○ Steadily Implement Talent Development Policies Based on AI Strategy 2022
○ Steadily implement talent development policies based on AI Strategy 2022, and consider further strengthening talent-related policies to develop internationally competitive talent not only in the development stage but also in the utilization stage of the foundational model era.
○ Establish an "AI Hub" that collects AI-related information and serves as a bridge between companies facing challenges and excellent technology/researchers. Also support community formation.
Bias in the data used for learning is an important challenge for foundational model AI. For example, some overseas image generation AI services cannot effectively generate images related to Japan.
The AI White Paper - Japan's National Strategy in the New Era of AI - (Summary Version)
*Generated by a non-Japanese generative AI platform using "Japanese shrine" as the prompt.
(3) Strengthening and utilizing computing resources
Building foundational model AI requires enormous computing resources, and the cost is immense.
Details
Recommendations ○ Referring to initiatives such as AIST's "AI Bridging Cloud Infrastructure," work on developing and expanding domestic infrastructure for the computing resources needed to build and utilize foundational model AI, and establish a new framework that relevant government agencies and private companies can share and utilize.
○ In light of the potential for further use of edge computing and the need to ensure stable access to computing resources for AI, promote the development of the semiconductor industry. Specifically, strengthen support for the design capabilities and research and development of high-performance semiconductors, for which rapid expansion of demand is expected in the future.
The use of AI in the administrative field brings social benefits such as improved quality and efficiency of government services. The government's promotion of AI use also stimulates local governments and private companies.
Recommendations ○ Investigate AI utilization cases and usage guidelines in government agencies of other countries, and use this information to plan and implement AI adoption in Japan.
○ It is important to immediately start multiple pilot projects that can show results in a short period as specific examples of AI using foundational models in government services.
○ Conduct hackathons and business contests to discover AI utilization projects in the administrative field.
○ Formulate Guidelines That Promote Thorough Utilization
○ Formulate guidelines that promote the thorough utilization of various AI technologies, including foundational models, in the administrative field.
○ Establish an AI implementation support team within the government, accumulate and analyze AI utilization cases, and support AI adoption in related agencies and organizations.
Examples of AI pilot projects in government: ○ Administrative work ensuring consistency with previously accumulated documents, such as drafting Diet responses, legal review support, analysis of government statistics, and preparation of meeting minutes.
○ Tasks such as checking incomplete application documents and responding to public inquiries about regulations and systems.
(2) Promoting AI Smart Cities Utilizing National Strategic Special Zones
Recommendations ○ The government strongly supports the efforts of local governments toward promoting AI-powered smart cities, utilizing the National Strategic Special Zones system.
○ Furthermore, review the current systems and operations of Super City-type National Strategic Special Zones and Digital Rural Fishing Villages Special Zones from the perspective of whether they are suitable for AI utilization, and make improvements promptly as necessary.
Private Companies in Japan, Including SMEs, Tend to Lag Behind in AI Adoption
Private companies in Japan, including SMEs, tend to lag behind in AI adoption. All companies need to seriously reconsider the impact of the AI era on their own businesses.
Recommendations ○ Conduct an urgent survey on the impact of foundational model AI on domestic industries.
○ Encourage the creation of various startups and new businesses utilizing AI. In particular, promote and support the cloudification of IT systems, which is a prerequisite for SMEs to benefit from AI's productivity improvements and other benefits.
○ Recommend that private companies and public institutions of a certain size appoint a Chief Digital Officer (CDO) responsible for AI utilization and data management.
○ It is necessary to discuss and formulate AI governance guidelines to promote not only risk management but also creativity and innovation, as appropriate.
○ Support measures to promote the utilization and treatment of AI human resources in companies, including upskilling, in order to support talent development efforts suited to the AI era.
(1) Regulatory Consideration in High-Risk Areas: Three Types of Critical AI Risks:
(1) Regulatory Consideration in High-Risk Areas: Three Types of Critical AI Risks:
- Risks of violating human rights, health, and safety
- National security risks from AI
- Risks of AI unduly interfering with democratic processes
There are no clear legal regulations on AI in Japan, but other countries are considering them.
Recommendations ○ Analyze AI regulation situations in foreign countries such as the EU, the US, and China, and conduct specific discussions — including potential legal regulation — on areas where legal measures are required in the new AI era, such as human rights violations, national security, and interference with democratic processes.
○ Japan should actively and strategically participate in discussions on international AI rules at various international forums, including the G7 Summit where Japan serves as chair this year, and cooperate with other countries to establish a global framework for AI use.
(2) An Approach to Regulations That Flexibly Responds to the New AI Era
Recommendations ○ Establish a mechanism that broadly shares information on technical verification of AI's potential uses across ministries and private companies, promoting the revision of analog regulations based on digital principles.
○ Regulatory Reform Council, Regulatory Sandbox
○ Improve the speed and accessibility of existing regulatory reform procedures such as the Regulatory Reform Council, Regulatory Sandbox, and the Gray Zone clarification process, and develop an environment where new businesses can be challenged without being bound by existing regulations.
○ Discuss the interpretation of intellectual property law related to generative AI, and consider formulating guidelines, in order to prevent abuse while promoting AI technology advancement and enabling further development of Japan's content industry.
AI has a major impact on the education sector
- AI will diversify learning methods
- Cultivating AI-native human resources is becoming increasingly important
- Regulation of AI use for problem-solving at schools is needed
Recommendations ○ The government should, in anticipation of the arrival of an AI-native era in which AI is actively utilized in everyday socioeconomic activities, position the improvement of AI literacy in public education curricula as a concrete skillset.
○ To prepare for an AI-native era in which AI literacy improvement is integrated into public education curricula and AI is actively utilized in everyday socioeconomic activities, it is important and should be done promptly to establish guidelines on the handling of AI in classrooms, including whether the use of LLMs is permitted.
This AI column is produced by the online assistant service "TIMEWELL."
Related Articles
- From full-time to part-time: the reality of working through two maternity leaves and how it changed my perspective on work | TIMEWELL
- Before taking parental leave: the 3 must-do points to ensure you can take leave even during busy season
- Embracing the role of a hands-on certified architect: finding my own path as the 5th generation of Fujita Construction