UTokyo Matsuo Lab, PKSHA, and Anthropic to Build 'Japan AI Index' to Visualize AI's Impact on Employment, Industry, Economy, and Education
The University of Tokyo's Matsuo Lab, PKSHA Technology, and Anthropic have announced a collaboration to build the 'Japan AI Index,' a foundational platform for continuously observing and analyzing the social impact of generative AI in Japan. It will integrate LLM usage statistics with domestic economic data to analyze AI's effects on employment, industry, and education.
📋 Article Processing Timeline
- 📰 Published: June 4, 2026 at 21:30
- 🔍 Collected: June 4, 2026 at 12:51
- 🤖 AI Analyzed: June 6, 2026 at 22:43 (57h 51m after Collected)
The Matsuo & Iwasawa Laboratory (led by Professor Yutaka Matsuo, hereinafter 'Matsuo Lab') of the Graduate School of Engineering at the University of Tokyo, together with PKSHA Technology Inc. (headquartered in Bunkyo-ku, Tokyo; CEO: Katsuya Uenoyama; hereinafter 'PKSHA') and Anthropic, PBC (hereinafter 'Anthropic'), announce a collaboration to build the 'Japan AI Index' (*1), a platform for continuously observing and analyzing the social impact of generative AI in Japan.
As Japan's working-age population continues to decline, leveraging AI to enhance productivity is an unavoidable challenge for maintaining the nation's industrial competitiveness. To appropriately promote the use of AI, it is necessary to understand, based on evidence, how AI adoption is spreading and what impact it is having on work styles, employment, and industry. The Japan AI Index is an advanced observation platform built by combining the academic neutrality of a university, statistical data on LLM usage from an LLM developer, and the on-the-ground knowledge of an industrial implementation company.
Background of the Japan AI Index:
Urgent need to build a foundation for evidence-based discussion amid rapid AI adoption.
With the proliferation of generative AI, its use is accelerating across all industries and occupations. To maximize its benefits, it is critical to objectively understand AI's impact on work styles, employment, and industry, and to reflect this in management and human resource development. The Japanese government has also established an AI Strategy Headquarters and an expert panel in 2025, aiming to become 'the country where AI is easiest to develop and utilize in the world' (*2).
However, there is currently no observation platform in Japan that can provide data on 'how AI is affecting society' while considering Japan's unique industrial structure and employment practices. In the United States, efforts are already underway. Anthropic has published the 'Anthropic Economic Index' (*3), which analyzes AI usage patterns at the occupation and task level based on anonymized Claude usage data, respecting user privacy.
Such an observation platform is crucial not only for objectively understanding the state of AI utilization in Japan but also for grasping its potential impact on Japanese society. The Japan AI Index will refine this knowledge according to Japan's industrial and employment statistics, building a unique observation platform for Japan.
Overview of the Japan AI Index:
Visualizing AI's impact by industry, occupation, and task by integrating LLM usage statistics with domestic economic data.
The Japan AI Index is an observation platform that integrates two streams of data. Its key feature is the unification of academic, technical, and industrial perspectives into a single index. The Matsuo Lab will lead the neutral analysis design. Anthropic will provide anonymized and aggregated Claude usage statistics, which are also used in its Economic Index, as foundational data for analysis. PKSHA will provide on-the-ground knowledge of AI implementation in the industry, helping to cultivate the index into a practical de facto standard.
1. Statistical data on LLM usage (including the Anthropic Economic Index)
2. Official statistics and survey data on Japan's economic activity, employment, and education (e.g., industry-specific productivity data, occupational data like Japan O*NET, employment figures)
By cross-referencing these data, the following analyses will be continuously conducted and published based on academic methods:
- Visualization of AI utilization progress in various Japanese industrial sectors
- Changes in the division of roles between tasks performed by AI and those performed by humans
- Examination of the relationship between AI adoption and GDP, employment, and wages
- Detection of productivity changes by industry and occupation
- Insights into the human resources and skills required in the AI era
The analysis results will be published by the University of Tokyo as an analysis dashboard and annual report, serving as an observation platform for discussions in policy, industry, and education. This will support decision-making in corporate AI investment, organizational and business design, and curriculum development at universities and educational institutions.
Visualization of AI utilization progress per working-age population estimated from Claude usage share. Usage is concentrated in the Tokyo, Osaka, and Kanagawa metropolitan areas. (Data from the Anthropic Economic Index as of February 2026)
International comparison of AI usage by occupation. Even in 'Computer and Mathematical' occupations, where adoption is most advanced, it is at 40% of the US level. (Data from the Anthropic Economic Index as of February 2026)
▼ Roles of Each Organization
Organization / Main Role in this Collaboration
University of Tokyo, Matsuo & Iwasawa Lab: Lead analysis entity. Responsible for observation platform design, statistical analysis, and issuing the analysis dashboard and annual report. Ensures neutrality from an academic standpoint.
Anthropic: Provides Claude free of charge for research support of the Japan AI Index, provides statistical data on Claude usage, and shares knowledge regarding the Anthropic Economic Index.
PKSHA Technology: Connects on-the-ground knowledge from the industry based on over 4,600 implementation cases. Promotes expansion of participating companies and manages planning and operations for the social penetration of the Japan AI Index.
*Specific forms of collaboration are scheduled to be finalized through discussions among the parties involved.
Future Outlook: Establishing a common language for AI utilization in Japan and promoting human-AI collaboration.
The first report and dashboard are scheduled for release in the fall of the 2026 academic year. Subsequently, while expanding the scope of analysis and recruiting participating companies (*4), information will be disseminated regularly through annual reports and quarterly updates.
Through this project, the Matsuo Lab, PKSHA, and Anthropic will visualize, with objective data, how AI is transforming work styles in various industries. They will create an environment that shifts discussions on policy, management, and human resource development to be 'evidence-based.' The goal is to build the foundational data for a rich society where AI expands human capabilities and possibilities, and to establish the Japan AI Index as a widely referenced observation platform.
Comments from Representatives of Each Organization:
[Professor Yutaka Matsuo, Matsuo & Iwasawa Laboratory, Graduate School of Engineering, University of Tokyo]
'With the rapid spread of generative AI, Japan still lacks a platform that can show with data what AI is actually changing. I believe the most important thing now is to create a state where we can discuss based on reality, not on feelings or impressions.
The Japan AI Index is a platform for continuously understanding how AI is penetrating various industries and occupations by cross-referencing Claude usage statistics with domestic employment and industrial data. Whether it's policy discussions, corporate management decisions, or the direction of human resource development, none can proceed with conviction without reliable data. This effort...'
As Japan's working-age population continues to decline, leveraging AI to enhance productivity is an unavoidable challenge for maintaining the nation's industrial competitiveness. To appropriately promote the use of AI, it is necessary to understand, based on evidence, how AI adoption is spreading and what impact it is having on work styles, employment, and industry. The Japan AI Index is an advanced observation platform built by combining the academic neutrality of a university, statistical data on LLM usage from an LLM developer, and the on-the-ground knowledge of an industrial implementation company.
Background of the Japan AI Index:
Urgent need to build a foundation for evidence-based discussion amid rapid AI adoption.
With the proliferation of generative AI, its use is accelerating across all industries and occupations. To maximize its benefits, it is critical to objectively understand AI's impact on work styles, employment, and industry, and to reflect this in management and human resource development. The Japanese government has also established an AI Strategy Headquarters and an expert panel in 2025, aiming to become 'the country where AI is easiest to develop and utilize in the world' (*2).
However, there is currently no observation platform in Japan that can provide data on 'how AI is affecting society' while considering Japan's unique industrial structure and employment practices. In the United States, efforts are already underway. Anthropic has published the 'Anthropic Economic Index' (*3), which analyzes AI usage patterns at the occupation and task level based on anonymized Claude usage data, respecting user privacy.
Such an observation platform is crucial not only for objectively understanding the state of AI utilization in Japan but also for grasping its potential impact on Japanese society. The Japan AI Index will refine this knowledge according to Japan's industrial and employment statistics, building a unique observation platform for Japan.
Overview of the Japan AI Index:
Visualizing AI's impact by industry, occupation, and task by integrating LLM usage statistics with domestic economic data.
The Japan AI Index is an observation platform that integrates two streams of data. Its key feature is the unification of academic, technical, and industrial perspectives into a single index. The Matsuo Lab will lead the neutral analysis design. Anthropic will provide anonymized and aggregated Claude usage statistics, which are also used in its Economic Index, as foundational data for analysis. PKSHA will provide on-the-ground knowledge of AI implementation in the industry, helping to cultivate the index into a practical de facto standard.
1. Statistical data on LLM usage (including the Anthropic Economic Index)
2. Official statistics and survey data on Japan's economic activity, employment, and education (e.g., industry-specific productivity data, occupational data like Japan O*NET, employment figures)
By cross-referencing these data, the following analyses will be continuously conducted and published based on academic methods:
- Visualization of AI utilization progress in various Japanese industrial sectors
- Changes in the division of roles between tasks performed by AI and those performed by humans
- Examination of the relationship between AI adoption and GDP, employment, and wages
- Detection of productivity changes by industry and occupation
- Insights into the human resources and skills required in the AI era
The analysis results will be published by the University of Tokyo as an analysis dashboard and annual report, serving as an observation platform for discussions in policy, industry, and education. This will support decision-making in corporate AI investment, organizational and business design, and curriculum development at universities and educational institutions.
Visualization of AI utilization progress per working-age population estimated from Claude usage share. Usage is concentrated in the Tokyo, Osaka, and Kanagawa metropolitan areas. (Data from the Anthropic Economic Index as of February 2026)
International comparison of AI usage by occupation. Even in 'Computer and Mathematical' occupations, where adoption is most advanced, it is at 40% of the US level. (Data from the Anthropic Economic Index as of February 2026)
▼ Roles of Each Organization
Organization / Main Role in this Collaboration
University of Tokyo, Matsuo & Iwasawa Lab: Lead analysis entity. Responsible for observation platform design, statistical analysis, and issuing the analysis dashboard and annual report. Ensures neutrality from an academic standpoint.
Anthropic: Provides Claude free of charge for research support of the Japan AI Index, provides statistical data on Claude usage, and shares knowledge regarding the Anthropic Economic Index.
PKSHA Technology: Connects on-the-ground knowledge from the industry based on over 4,600 implementation cases. Promotes expansion of participating companies and manages planning and operations for the social penetration of the Japan AI Index.
*Specific forms of collaboration are scheduled to be finalized through discussions among the parties involved.
Future Outlook: Establishing a common language for AI utilization in Japan and promoting human-AI collaboration.
The first report and dashboard are scheduled for release in the fall of the 2026 academic year. Subsequently, while expanding the scope of analysis and recruiting participating companies (*4), information will be disseminated regularly through annual reports and quarterly updates.
Through this project, the Matsuo Lab, PKSHA, and Anthropic will visualize, with objective data, how AI is transforming work styles in various industries. They will create an environment that shifts discussions on policy, management, and human resource development to be 'evidence-based.' The goal is to build the foundational data for a rich society where AI expands human capabilities and possibilities, and to establish the Japan AI Index as a widely referenced observation platform.
Comments from Representatives of Each Organization:
[Professor Yutaka Matsuo, Matsuo & Iwasawa Laboratory, Graduate School of Engineering, University of Tokyo]
'With the rapid spread of generative AI, Japan still lacks a platform that can show with data what AI is actually changing. I believe the most important thing now is to create a state where we can discuss based on reality, not on feelings or impressions.
The Japan AI Index is a platform for continuously understanding how AI is penetrating various industries and occupations by cross-referencing Claude usage statistics with domestic employment and industrial data. Whether it's policy discussions, corporate management decisions, or the direction of human resource development, none can proceed with conviction without reliable data. This effort...'
FAQ
What is the purpose of the Japan AI Index?
To visualize the impact of AI on Japan's employment, industry, and education based on evidence, supporting policy and management decisions.
Who is leading this project?
The University of Tokyo's Matsuo & Iwasawa Lab leads the analysis design, with Anthropic and PKSHA Technology providing data and expertise.
When will the first output be released?
The first report and analysis dashboard are scheduled for release in the fall of the 2026 academic year.