The Group of Seven (G7) industrial countries has announced a new set of guidelines, the International Code of Conduct for Organizations Developing Advanced AI Systems, for organizations developing advanced artificial intelligence (AI) technologies. The code, which builds upon the “Hiroshima AI Process“, was made public on the same day that U.S. President Joe Biden issued an Executive Order on “Safe, Secure and Trustworthy AI”.
This global AI development framework aims at promoting trustworthy, safe, and secure AI. It comes in tandem with major AI developments such as the finalizing of the European Union's financially binding EU AI Act and the United Nations Secretary-General's new Artificial Intelligence Advisory Board.
The centerpiece of the new code of conduct is a comprehensive 11-point framework for responsible AI creation and enterprise, including guidelines for identifying, evaluating, and mitigating risks during AI development and deployment. It extends the Hiroshima Process, which was announced in May. This agreement seeks to ensure that AI development and deployment align with the shared democratic values of the G-7 nations.
A Comprehensive 11-Point Framework for AI Development
The 11-point framework is designed to guide AI developers with stringent analytics and transparent reporting measures to fortify AI integrity. It includes proactive strategies for identifying and mitigating potential advanced AI vulnerabilities and regularly monitoring its usage after deployment.
Developers are also encouraged to publicly disclose the capabilities and limitations of advanced AI systems, along with domains of appropriate and inappropriate use. The framework also calls for the development and adoption of international technical standards.
Among other concerns, deploying reliable content authentication mechanisms, advancing research to mitigate societal, safety and security risks, and implementing appropriate measures to protect personal data and intellectual property are amongst the vital principles outlined in the code.
Unifying Global Efforts Towards Responsible AI Growth
The initiative received global government approval and has been lauded for its efforts to unionize global approaches towards AI development with due regard for law, human rights, due process, diversity, fairness, non-discrimination, democracy, and “humancentricity”.
In a unified voice, the G7 countries have urged organizations worldwide dealing with AI to commit to this code of conduct. They also acknowledged that “different jurisdictions may take their own unique approaches to implementing these guiding principles“. The G7 nations, namely United States, Britain, European Union, Canada, France, Germany, Italy, and Japan aim to ensure that AI is developed responsibly and does not pose a substantial risk to security, safety, and human rights.
In a bid to encourage AI development for global societal issues like climate crisis, global health and education, the G7 nations have emphasized the potential benefits while ensuring that the risks are mitigated. An environment fostering AI benefits, closing digital divides and achieving digital inclusion is the ultimate aim behind this global initiative.
A Growing Push for Meaningful AI Regulation
Governments worldwide, recognizing the transformative potential of AI, are grappling with the challenge of regulating generative AI. The European Union, for instance, has put forth its proposed AI Act, which places a strong emphasis on transparency rules for foundation models. Companies such as Microsoft's GitHub and Hugging Face have called for the AI Act to be open-source friendly, while OpenAI has argued the proposed laws are too strict.
In the UK, the Competition and Markets Authority (CMA) has recently took a lead in AI regulations by unveiling a comprehensive set of principles aimed at guiding the development and deployment of AI foundation models. And in the United States, the Biden Administration recently announced eight more tech companies, including Adobe, IBM, Nvidia, Cohere, Palantir, Salesforce, Scale AI, and Stability AI, have pledged their commitment to the development of safe, secure, and trustworthy artificial intelligence (AI). This move builds upon the Biden-Harris Administration's efforts to manage AI risks and harness its benefits.
In July when the initiative was launched, Leading U.S. tech companies, including OpenAI, Google, Microsoft, Amazon, Anthropic, Inflection AI, and Meta, agreed to the voluntary safeguard. The commitments are divided into three categories: Safety, Security, and Trust, and apply to generative models that surpass the current industry frontier.