As an OECD legal instrument, the principles represent a common aspiration for its adhering countries. The OECD AI Principles focus on how governments and other actors can shape a human-centric approach to trustworthy AI. During the development of the AI Principles, the OECD has actively conducted empirical research and participated in policy initiatives to contribute significantly to the global AI policy conversation. Additionally, the OECD has conducted thorough analysis and measurement activities to better understand the AI technical landscape, assess the socioeconomic impacts of AI technologies, identify key policy considerations, and document AI initiatives at both national and international levels. The work culminated in May 2019 when the OECD AI Principles were formally adopted. The OECD AI Principles provide a practical and adaptable framework promoting innovative and trustworthy AI. They emphasise the importance of upholding human rights and democratic values in the development and deployment of AI technologies. Central to the principles is the call for a human-centric approach, with governments and stakeholders playing a pivotal role in shaping AI policy.
The OECD AI Principles are not legally binding, but rather, represent a collective commitment among member countries to maintain fundamental standards in the realm of AI. The scope of the AI Principles covers AI actors. The AI Actors encompass organisations and individuals that are actively involved in the entire lifecycle of an AI system, including those who deploy or operate AI technologies.
The OECD AI Principles advocate for AI that combines innovation with trustworthiness while upholding human rights and democratic principles. The following AI Principles form the core of the OECD guidance: Inclusive growth, sustainable development and well-being, human-centered values and fairness, transparency and explainability, robustness, security, and safety, as well as accountability. These principles provide clear directives for shaping public policies and strategies, and their broad applicability ensures their relevance in the context of AI advancements globally.
