Messages from Keidanren Executives and Contributed articles to Keidanren Journals April, 2024 How International Standards can accelerate responsible AI innovation.
The risks and opportunities of AI
We live in a fast-moving, interconnected world, with new technologies changing the way people do business, collaborate and consume information. Artificial intelligence (AI) is the most prominent example of this, powering everything from text and image generation to music and movie recommendations, while creating headlines across both scientific media and the mainstream press.
AI has the potential to revolutionize our societies and economies, with the benefits including game-changing advances in technology, healthcare and medicine. As the world's leading standards organization, ISO is ready to develop the appropriate tools to respond to these future needs.
However, while the capabilities of AI grow exponentially, there are deep concerns about privacy, bias, inequality, safety and security. As the pace of technological advancement in the field quickens and regulation struggles to keep up, International Standards that help mitigate the risks and maximize the rewards can act as a framework upon which AI systems are developed, used and regulated. They underpin and complement regulation, providing appropriate guardrails for responsible, safe and trustworthy AI development.
By focusing on interoperability, transparency and safety, International Standards are creating a universal language around AI that brings together countries, industries and users and promotes trust.
The benefits of the ISO system
Thanks to the transparency and openness of ISO's consensus-based standards development process, our standards reflect the needs of a diverse range of global stakeholders, including business, industry and regulators, addressing the multiple ways in which AI can be used. The ISO system provides a neutral platform and is open to everyone by means of engagement with our member bodies.
Our AI technical community boasts participation from 60 countries, demonstrating the breadth of perspectives within the standards it develops. It also engages with several major actors, including UNESCO and OECD, and conducts workshops which offer collaboration opportunities to a wide range of stakeholder groups.
ISO is committed to engaging with business, industry, regulators, policymakers, international organizations and end users in order to develop standards that will ensure the widest, most equitable and responsible adoption of AI.
For this reason, ISO leaders convened with prominent speakers from business and international organizations at the World Economic Forum's recent Annual Meeting in Davos. They participated in several sessions on AI, fostering valuable discussions and insights and making the case for International Standards as solutions to the various challenges posed by AI.
A framework for AI governance
Looking at how AI risk impacts users is crucial to ensuring the responsible and sustainable deployment of AI technologies. More than ever, businesses today need a framework to guide them on their AI journey. ISO/IEC 42001, the world's first AI management system standard, meets that need.
This groundbreaking standard provides guidelines for the governance and management of AI technologies. It offers a systematic approach to addressing the challenges associated with AI implementation in a recognized management system framework, covering areas such as responsible AI, accountability, transparency and data privacy.
Walking the tightrope between opportunity and risk is only possible with a robust governance in place. This is why it is important for business leaders to educate themselves on ISO/IEC 42001. With the rapid uptake of AI worldwide, this standard is set to become an integral part of every organization's success, following in the footsteps of other management systems standards such as ISO 9001 for quality, ISO 14001 for environment and ISO/IEC 27001 for IT security.
In addition, ISO, in collaboration with the International Electrotechnical Commission (IEC), has developed a number of innovative standards covering the entire AI ecosystem. To give a couple of examples, ISO/IEC 23894 focuses on the management of risk in AI systems, thereby building trust and confidence. ISO/IEC 22989, meanwhile, establishes terminology and describes concepts in the field of AI.
This group of horizontal AI standards can be applied to specific domains across the ISO ecosystem; any technical committee can therefore produce sector-specific applications.
Conclusion
As AI becomes more powerful and pervasive, we must ensure it is developed and used responsibly, addressing issues of bias, privacy and transparency. International Standards help to achieve this, striking a balance between risk and opportunity. If we approach AI with a positive mindset, placing societal needs such as ethics and sustainability at the heart of its development, then we can accelerate responsible AI innovation and unlock the full potential of this transformational technology.