Hiroshima Process International Guiding Principles for Organizations Developing Advanced AI System
The Hiroshima Process International Guiding Principles provide a non-exhaustive, living framework to promote safe, secure, and trustworthy advanced AI systems globally. Developed to build on existing OECD AI Principles, they offer guidance for organizations across academia, civil society, and the private and public sectors. The principles cover risk management, safety testing, incident reporting, transparency, cybersecurity, content provenance, and responsible AI research.
Summary and content fields are drawn from the document text and are pending editorial review.
Who issued it
Who it governs
Organizations developing advanced AI systems should read this to understand international expectations for safe, secure, and trustworthy AI governance.
What the document contains
The document addresses risk management, safety testing, post-deployment monitoring, transparency reporting, incident sharing, governance policies, security controls, content provenance, AI safety research, global challenge applications, standards development, and data protection for advanced AI systems.
The document does not provide binding legal enforcement mechanisms, statutory penalties, or specific technical definitions for advanced AI thresholds.
- Organizations developing advanced AI systems should take appropriate measures to identify, evaluate, and mitigate risks across the entire AI lifecycle.
- Developers should monitor systems after deployment, manage vulnerabilities, and facilitate reporting of incidents and issues.
- Transparency should be supported through public reporting of capabilities, limitations, and transparency reports for new releases.
- Organizations need to implement robust security controls, including cybersecurity, physical security, and insider threat safeguards.
- Content authentication, provenance mechanisms, and watermarking should be utilized where technically feasible to identify AI-generated content.
· 1. Identify, evaluate, and mitigate risks across the AI lifecycle · 2. Monitor for vulnerabilities and misuse after deployment · 3. Publicly report capabilities, limitations, and transparency information · 4. Work towards responsible information sharing and incident reporting · 5. Develop and disclose AI governance and risk management policies · 6. Invest in and implement robust security controls · 7. Develop and deploy reliable content authentication and provenance mechanisms · 8. Prioritize research to mitigate societal, safety, and security risks · 9. Prioritize the development of advanced AI systems to address global challenges · 10. Advance the development and adoption of international technical standards · 11. Implement appropriate data input measures and protections for personal data and intellectual property
How it scores
Each score is backed by a quote verified against the document text. Editorial review has not yet confirmed them.
Accountability
4/5Clear responsibility for AI outcomes
This should include accountability and governance processes to evaluate and mitigate risks, where feasible throughout the AI lifecycle.
Principle 5
screening score · 2026-10-25
Fairness
3/5Non-discrimination and equitable treatment
Organizations are encouraged to take appropriate measures to manage data quality, including training data and data collection, to mitigate against harmful biases.
Principle 11
screening score · 2026-10-25
Human Dignity
3/5Respect for the inherent worth of persons
While harnessing the opportunities of innovation, organizations should respect the rule of law, human rights, due process, diversity, fairness and non -discrimination, democracy, and human- centricity
Paragraph 6
screening score · 2026-10-25
Human Oversight
1/5Humans remain in control of consequential decisions
Organizations should prioritize responsible stewardship of trustworthy and human-centric AI and also support digital literacy initiatives.
Principle 9
screening score · 2026-10-25
Privacy
4/5Protection of personal data and private life
Develop, implement and disclose AI governance and risk management policies, grounded in a risk -based approach – including privacy policies, and mitigation measures
Principle 5
screening score · 2026-10-25
Transparency
4/5Openness about how AI systems work and are used
Publicly report advanced AI systems’ capabilities, limitations and domains of appropriate and inappropriate use, to support ensuring sufficient transparency
Principle 3
screening score · 2026-10-25
Enforcement
1/5Consequences and compliance mechanisms with real force
We also commit to develop proposals, in consultation with the OECD, GPAI and other stakeholders, to introduce monitoring tools and mechanisms to help organizations stay accountable
Paragraph 5
screening score · 2026-10-25
Impact Assessment
4/5Required assessment of an AI system's risks/impacts before or during use
Take appropriate measures throughout the development of advanced AI systems, including prior to and throughout their deployment and placement on the market, to identify, evaluate, and mitigate risks across the AI lifecycle.
Principle 1
screening score · 2026-10-25