Hitachi and Columbia University Publish Joint Report on Opportunities and Risks of AI for the Transition to a Sustainable Society

Hitachi, together with the Columbia Center on Sustainable Investment (CCSI) at the Columbia Climate School, has published a joint research report titled "Artificial Intelligence and Sustainability Transitions: Emergent Opportunities, Risks, and Governance" in English.

This report examines how AI influences the transition to a sustainable society, organizing and analyzing opportunities and risks across five domains: "Planetary Boundaries," "Energy Systems," "Industry and Labor," "Finance," and "Democracy and Social Resilience." Based on this analysis, it proposes a governance framework to harness the benefits of AI while mitigating its risks, along with a roadmap for its implementation. A Japanese version is scheduled for release in the future.

Through its social innovation business leveraging IT, OT, and products, Hitachi has been creating value from data with Lumada at its core, working to solve challenges for customers and society. The use of AI is an indispensable element in further evolving these efforts. At the same time, Hitachi aims to realize a Harmonized Society where the environment, well-being, and economic growth are in harmony, by not only enjoying the opportunities of AI utilization but also confronting its risks. Based on the insights gained from this joint research, Hitachi will contribute to the realization of a sustainable and resilient society through the balance of responsible AI utilization and governance.

Research Background

Against the backdrop of intensifying climate change, geopolitical tensions, and resource constraints, achieving a sustainable society has become more difficult. Furthermore, AI technologies, including generative AI, are rapidly evolving and bringing about significant changes across a wide range of societal fields. While AI has the potential to accelerate the transition to a sustainable society, it also creates new challenges such as massive energy consumption and impacts on employment.

Amidst this, there is a growing movement to re-examine AI from a sustainability perspective, and international discussions, particularly around "AI governance," are intensifying. Initiatives such as the Independent International Scientific Panel on Artificial Intelligence (IISP-AI), to be held at the United Nations in July 2026, are advancing efforts to scientifically understand AI's capabilities and risks and to form international rules. However, research that cross-sectionally analyzes the relationship between "AI and sustainability" across diverse fields has been insufficient. Therefore, Hitachi, which advocates for the realization of a Harmonized Society and has supported social infrastructure through AI technology, and CCSI, which has been conducting research at Columbia University—home to the first graduate school in the United States specializing in climate change—undertook this joint research to comprehensively examine the impact of AI development on society as a whole.

Report Summary

This research is a joint exploratory study conducted by Hitachi and CCSI as part of "Transitions Research," which explores pathways for transformation towards a sustainable future. The main contents of the report are as follows:

1. Systematic Organization of AI's Impact on a Sustainable Society Across Five Domains

The impact of AI on a sustainable society was organized and analyzed from five domains: "Planetary Boundaries," "Energy Systems," "Industry and Labor," "Finance," and "Democracy and Social Resilience." The report shows that AI has the potential to accelerate the transition to a sustainable society through environmental monitoring, advanced energy operations, improved industrial productivity, and expanded use of information in finance. At the same time, it indicates that because the manifestation of these impacts varies by domain, they must be understood as cross-cutting changes rather than isolated technological issues.

2. AI Simultaneously Creates Contributions to Sustainability and New Risks

This research positions AI not simply as a "technology useful for sustainability" but as a foundational technology that presents both significant opportunities and serious risks. For example, while AI can contribute to the integration of renewable energy and early detection of environmental risks, it also creates new negative externalities. These include the power and water consumption of data centers, access disparities, the spread of disinformation, and risks from future advanced AI use that could cause individual problems to cascade and amplify across multiple domains, increasing overall societal instability and vulnerability. The report demonstrates that the value of AI is determined not by the technology itself, but by how it is designed, implemented, and governed.

3. Three-Tier Governance is Necessary to Harness AI's Benefits and Mitigate Risks

To address the impacts of AI, the report demonstrates the necessity of a three-tier governance structure: sectoral, cross-domain, and global. Sectoral governance requires establishing rules and oversight tailored to the context of each field, such as environment, labor, finance, and democracy. Cross-domain governance calls for common safeguards like transparency, data quality, auditing, and human oversight. Furthermore, to enable countries to cooperatively address the risks that AI development poses to society and the economy, the report recommends that a global governance framework is indispensable.

4. Presentation of a Phased International Roadmap for AI Governance

To make AI governance effective, the report presents a three-phase international roadmap. Phase 1 involves forming a common scientific understanding under the UN framework. Phase 2 involves establishing a provisional international safety framework for high-risk areas. Phase 3 proposes evolving into an international framework convention where countries coordinate implementation methods while assuming common obligations. This provides a pathway to align AI development with the long-term well-being of the planet and its people.

5. Analysis Based on Literature Review and Expert Insights

The analysis is based not only on a review of existing literature but also on interviews with experts from international organizations and universities. This incorporates multifaceted perspectives that go beyond mere technological discussion. Experts who contributed to the interviews include Tshilidzi Marwala, Rector of the United Nations University (UN Under-Secretary-General), as well as AI specialists from the International Labour Organization (ILO), University College London, the University of Oxford, and the University of Waterloo.

Future Outlook

Hitachi will share the insights gained from this research internally and externally, and will advance research and development aimed at accelerating the construction of a sustainable world using AI.