Page 25 - AI for Good - Impact Report

AI for Good

The UN’s work, supported by the ITU, focuses on embedding ethical considerations and fostering
international collaboration, all while advancing the SDGs. The G7 emphasizes creating a flexible
framework for transparency and accountability, while the Bletchley Declaration promotes
cooperative safety measures for frontier AI technologies. Complementing these efforts, the
OECD's 2023 definition of AI systems may provide a foundational framework that supports
harmonization among national policies and contributes to the development of cohesive global
AI governance.

Standard setting

The development of AI policies and regulations, including frameworks and the standardization
of AI, are closely interconnected, with each process reinforcing and complementing the other.
Standards provide the technical specifications and best practices that are crucial for the creation
of robust regulatory and policy frameworks and a means of assuring that those regulatory
requirements are being met. They lay the groundwork for ensuring interoperability between AI
systems. Conversely, policy frameworks and regulations shape the evolution of these standards,
ensuring that they adequately address legal, ethical, and societal concerns.

Given this concerted effort, it is essential to consider the technical foundations upon which
AI standards are built. Several organizations are actively engaged in the development of AI
standardization, working together to create comprehensive standards that span from broad
ethical guidelines to specific technical requirements. These standards are designed to be
applicable across a wide range of industries and technologies, ensuring that AI is implemented
safely and effectively.

When discussing cooperation on AI standardization, the World Standards Cooperation (WSC)
plays a pivotal role. The WSC is a collaborative initiative between the International Organization
for Standardization (ISO) , the International Electrotechnical Commission (IEC) , and the
96
97
International Telecommunication Union (ITU). Its mission is to enhance the global framework
for AI standards by advancing a consensus-based system that promotes the safe, responsible,
and effective use of AI across industries through international cooperation. Central to the WSC’s
efforts is fostering collaboration among international stakeholders, sharing expertise, and
shaping the future trajectory of AI governance. In line with these goals, the WSC recently issued
a call for collaboration on AI standards through the AI for Good initiative, urging governments,
industry, and other stakeholders to work together to establish robust standards for AI. Reflecting
its commitment to AI governance, the WSC will host the AI for Good International AI Standards
Summit in New Delhi, India, in October 2024. This summit will focus on multi-stakeholder
cooperation to develop global standards for AI watermarking, multimedia authenticity, and
deepfake detection technologies, further advancing the work of the WSC in these areas.

A key contributor to these efforts is the ITU, which has already published or is developing over
220 AI standards. These standards span a wide range of areas, providing essential guidelines
and frameworks to ensure the safe and ethical deployment of AI technologies. The ITU’s
standards work includes pre-standardization Focus Groups on topics like AI and Health (in
collaboration with World Health Organization), AI and Autonomous and Assisted Driving, AI
and Environmental Efficiency, AI and Natural Disaster Management (with World Meteorological
Organization and United Nations Environment Programme), AI and 5G, AI and Internet of
Things (IoT) for Digital Agriculture (with Food and Agriculture Organization), and the AI for
Road Safety Initiative.

15