G7 Hiroshima AI Process
October 2023 — Published. Voluntary Code of Conduct for AI developers.
medium risk framework
Organizations developing advanced AI systems in G7 countries (US, UK, EU, Canada, Japan, Germany, France, Italy). Also open for companies worldwide to voluntarily adopt the Code of Conduct.
Overview
The G7 Hiroshima AI Process was launched at the 2023 G7 Summit to develop international guiding principles for advanced AI, particularly foundation models. It produced the International Guiding Principles and a Code of Conduct for AI developers that align with safety, transparency, and trustworthiness goals.
Key Requirements
- Identify and mitigate risks across the AI lifecycle
- Report AI capabilities and safety to governments
- Invest in cybersecurity and insider threat safeguards
- Watermarking and labeling of AI-generated content
- Responsible disclosure of AI vulnerabilities
- Prioritize safety research and risk mitigation
- Implement technical safeguards against misuse
- Advance international technical standards for AI safety
Implementation Guidance
- 1Review and adopt the Hiroshima AI Code of Conduct voluntarily
- 2Align AI safety practices with the 11 guiding principles
- 3Ensure content labeling policies comply with watermarking principles
- 4Monitor G7 and GPAI (Global Partnership on AI) updates
Penalties for Non-Compliance
Voluntary framework — no direct penalties. Non-compliance may influence national regulatory actions.
Framework Details
Short Name
Hiroshima AI Process
Jurisdiction
G7 Nations
Status
Risk Level
Enforcement Date
October 2023 — Published. Voluntary Code of Conduct for AI developers.
Affected Organizations
Organizations developing advanced AI systems in G7 countries (US, UK, EU, Canada, Japan, Germany, France, Italy). Also open for companies worldwide to voluntarily adopt the Code of Conduct.
Tags
This is educational guidance only. Always consult qualified legal counsel for compliance decisions affecting your organization.