AI Ethics and
Governance

AI Ethics and Governance

Message from

The Chief
AI Governance Officer

Our lives and society are changing at an unprecedented pace as a result of technological innovation.

In particular, AI technologies, including generative AI, are significantly transforming the ways in which information is created, analyzed, and utilized, and are having a profound impact on business and society. AI is no longer merely a tool for improving efficiency; it is becoming an important foundation that supports decision-making and social infrastructure. Amid these changes, companies are highly expected not only to promote the use of AI, but also to take responsibility for its impact.

While advances in AI bring tremendous possibilities, they also give rise to new ethical and social challenges, such as data bias, transparency, privacy, and misinformation specific to generative AI. Addressing these challenges appropriately and using AI in a manner that earns the trust of society are essential elements for sustainable corporate growth.

Guided by our corporate philosophy of “Information Revolution — Happiness for everyone,” SoftBank Corp. is committed to maximizing the potential of AI while positioning the realization of responsible AI as a key management priority. To achieve this, we are developing AI ethics guidelines and strengthening our governance framework, while ensuring thorough risk management across the entire AI lifecycle, from development to operation.

In addition, while taking into account regulatory trends in Japan and abroad as well as societal expectations, we will further strengthen collaboration with governments, academic institutions, and partner companies, and contribute to building a foundation of trust across society.

By balancing the promotion of innovation with governance, we will continue working toward the realization of a society in which everyone can enjoy the benefits of AI with confidence. Through the continued practice of responsible AI, we aim to retain our position as a company that is trusted by society.

Tadashi Iida
Senior Vice President,
CISO (Chief Information Security Officer)
& CRO (Chief Risk Officer)
CAIGO (Chief AI Governance Officer)
SoftBank Corp.

Tadashi Iida Senior Vice President, CISO (Chief Information Security Officer) & CRO (Chief Risk Officer) CAIGO (Chief AI Governance Officer) SoftBank Corp. Tadashi Iida Senior Vice President, CISO (Chief Information Security Officer) & CRO (Chief Risk Officer) CAIGO (Chief AI Governance Officer) SoftBank Corp.

The Chair
of the AI Ethics Committee

Kyoko Hoashi Senior Director, Digital Governance Division, IT Strategy Division, Technology Unit Chair of the AI Ethics Committee SoftBank Corp. Kyoko Hoashi Senior Director, Digital Governance Division, IT Strategy Division, Technology Unit Chair of the AI Ethics Committee SoftBank Corp.

Advances in AI technology are significantly transforming the way we work and deliver services. In particular, with the widespread adoption of generative AI, it has become more important than ever to address risks such as misinformation, bias, and privacy concerns, alongside the convenience that AI provides.

At SoftBank Corp., we are developing AI ethics guidelines and strengthening our governance framework to support the appropriate use of AI. Across each process, from development to operation, we are promoting initiatives that emphasize fairness, transparency, and explainability, while managing risks throughout the entire AI lifecycle.

In addition, through employee education and awareness-raising activities, we are working to improve AI literacy and ethical awareness, while also pursuing continuous improvement in light of external trends and societal expectations.

We will continue to promote practical and effective AI governance rooted in the front lines of our business, and earn society’s trust by realizing AI that can be used with confidence.

Kyoko Hoashi
Senior Director, Digital Governance Division,
IT Strategy Division, Technology Unit
Chair of the AI Ethics Committee
SoftBank Corp.

Policy

With careful consideration of the ethical risks and social impacts associated with the use of AI, we are committed to ensuring that AI is utilized correctly and responsibly for the well-being of all people. To guide this vision, we have established our conceptual framework—the AI Ethics Policy. This policy is endorsed by the Board of Directors.

As a practical implementation of this policy, we have developed a range of supporting documents, including internal regulations, guidelines, and checklists. These resources are used across the organization to promote human-centric, fair, transparent, accountable, and other responsible AI practices. Through these efforts, we aim to ensure the proper use of technology in harmony with society, and to realize AI that is trusted by all.

1.AI Ethics Policy 2.Rules and Srandards 3.Guidelines 4.Check Sheet

AI Ethics and
Governance Framework

To comply with all relevant laws, guidelines, and social norms concerning the development and use of AI, and to ensure that AI is utilized appropriately for the well-being of people, we promote governance over internal departments that use AI through the Digital Governance Division, which is responsible for various forms of governance, including AI governance.

The Chief AI Governance Officer, who is responsible for AI governance at our company, is responsible for decision-making and management oversight related to AI governance. The officer also oversees the implementation status of AI governance in each department and helps ensure the effectiveness of company-wide AI governance through measures such as appointing AI Governance Managers, who are responsible for risk management and promoting improvements.

In addition, the AI Ethics Committee, composed of experts from both inside and outside the company, serves as an Advisory Board and provides guidance that contributes to the advancement of governance.

As part of our regular governance structure, the internal departments responsible for using AI are accountable for the validity of outcomes, risk decisions, and compliance assurance in each use of AI, and bear accountability within the company. These departments are responsible for appropriately managing the purpose of AI use, the validity of outcomes, risk assessment results, and compliance with laws, regulations, and internal rules, as well as reporting and explaining such matters to the Digital Governance Division and other relevant departments.

In cases where concerns arise regarding the inappropriate use of AI or its potential social impact, the Chairperson of the AI Ethics Committee collaborates with relevant department heads to promptly assess the situation, evaluate risks, and formulate and implement appropriate responses.

Moreover, in the event of a significant ethical or social incident involving AI, we will establish an Emergency Response Headquarters led by the President. This unit will coordinate with both internal and external stakeholders to ensure an effective response and, when necessary, report appropriately to relevant government ministries and agencies.

AI Ethics and Governance Framework

AI Ethics Committee

As the use of AI continues to expand across various industries, its applications are expected to become increasingly diversified and technically sophisticated. At the same time, AI remains a technology that requires careful ethical consideration. Depending on how it is utilized, AI may lead to discriminatory evaluations or biased selection processes.

To address these concerns, we operate the AI Ethics Committee, chaired by Kyoko Hoashi, Senior Director at the Digital Governance Division, which is responsible for various areas of governance, including AI, with participation from external experts. The committee engages in discussions and proposals on a range of ethical issues related to AI. Our goal is to realize objective and effective AI governance from a user-oriented perspective.

The AI Ethics Committee is responsible for the following:

  • Sharing and proposing policies, initiatives, and plans related to AI ethics
  • Disseminating information that supports ethical AI practices
  • Reviewing internal AI usage and consultation cases, identifying ethical issues, and offering advice

Supporting AI Ethics
and governance
across group companies

We provide support for AI ethics and governance to our affiliated companies, including subsidiaries and related entities, with the aim of reducing and proactively preventing risks associated with the use of AI technologies. This includes promoting information sharing on appropriate risk countermeasures.

In addition, many of our affiliated companies have adopted the SoftBank AI Ethics Policy as a guiding framework to ensure appropriate and consistent use of AI in both application and research and development activities.

Initiatives

As we continue to strengthen our AI ethics and governance framework, we conduct ethical reviews, provide expert guidance, and carry out risk assessments based on a risk-based approach during the development and deployment of AI technologies within the company. To continuously monitor and evaluate AI usage and its social impact, we have established an AI Governance Monitoring Function, which facilitates close collaboration with both internal and external stakeholders. These activities are conducted in reference to domestic and international guidelines, enabling us to revise countermeasures and explore new initiatives as needed. As one such initiative, we are advancing our efforts toward obtaining certification under ISO/IEC 42001, the international standard for AI management systems. As part of this process, we have undergone a certification audit and continue to take the necessary steps toward achieving certification.

However, simply establishing rules is not sufficient to embed desirable behavior across the organization. For rules to be effective, they must be fully understood by all stakeholders and naturally integrated into their daily work practices.

Therefore, in addition to developing robust governance rules, we place strong emphasis on education and training to communicate these principles clearly. We believe that only when each employee understands the context and intent behind the rules —and applies them in their own actions— can the governance truly function effectively.

Risk-based approach

We have adopted a risk-based approach for conducting AI ethics and governance risk assessments. Risk levels are defined using a four-tier scale: Prohibited, High, Middle, and Low. In defining these risk levels, we consider the potential social and business impact in the event that a risk materializes. At the same time, we ensure that the criteria are designed to be simple and clear, allowing employees to classify risks easily and consistently.

For important decisions involving AI, we verify whether the design allows human review or modification according to the level of risk. From the perspective of preventing misuse and abuse, we also confirm that the purpose of use and scope of functionality are clearly defined and that documentation is in place to encourage appropriate use. In addition, we use checklists and other tools to confirm the appropriateness of governance structures and authority management, the status of continuous monitoring of AI model performance and other factors, and whether bias has been identified and improvement measures have been implemented. The department in charge regularly reviews the implementation status of these measures, and the Digital Governance Division also periodically monitors the operational status.

Risk-based approach
[Note]
  • *
    Digital Governance Division

Education

We place strong emphasis on promoting AI ethics and governance education to ensure that internal rules and knowledge related to AI are properly recognized, put into practice, and firmly embedded across the organization. As of FY2025, we have implemented a comprehensive education program that includes:

  • Annual e-learning sessions
  • Biannual study sessions
  • Monthly email newsletters

These initiatives are directed at all employees, including executives. In addition to annual participation by all employees, e-learning is mandatory for new graduate employees and mid-career hires when they join the company.

Our educational content focuses on enhancing AI literacy across the organization. Topics include:

  • Case studies of AI-related incidents occurring domestically and internationally
  • Key considerations when using AI —including generative AI— such as bias, data leakage, copyright infringement, and hallucination risks
  • Updates on social trends and developments related to AI ethics
e-learning / Study Session / Email Newsletter Delivery

Collaboration
with stakeholders

We regularly share the details of our initiatives with relevant government ministries and organizations. These efforts have been highly regarded and are featured in the AI Guidelines for Business.

We also participate in the reporting framework (in Japanese) under the International Code of Conduct based on the Hiroshima AI Process, and are working to ensure and improve the safety and reliability of AI. We provide highly transparent disclosure of information on our AI-related initiatives through submitted reports and other means.

Going forward, we will continue to monitor developments in domestic and international regulations and guidelines, while pursuing practical and forward-thinking initiatives and ensuring their transparency through public disclosure.