This OECD Forum 2019 background note will be used to prepare speakers on the panel Artificial Intelligence: Next Steps, taking place at the OECD headquarters from 14:30-15:45 on Tuesday, 21 May. Join the Forum Network to comment and help inform the upcoming debate and, whether you're with us in Paris or watching online, let us know what you think of the session!
Artificial intelligence (AI) is rapidly permeating and transforming every aspect of our lives and in such a way that we barely notice it. It is present in our smart home devices, when we interact with customer services or as part of the recruitment screening when we apply for a job. We often depend on it to do certain tasks. Moreover, it is also influencing the way we work and opening new avenues to address global challenges like climate change and access to quality medical care.
AI represents a fundamental shift as, by its nature, it will affect every single sector in our economy and our societies. While it holds the promise to generate productivity gains, boost efficiency, lower costs and improve well-being, it also raises important questions about how to protect privacy and human rights, as well as to ensure the safety, transparency, explainability, inclusiveness and accountability of AI systems.
Still, if we aim to minimise the risks and maximise the opportunities of AI, no country or actor can effectively tackle this in isolation: as technology doesn't stop at borders, the global reach of AI requires a global approach, involving countries at all levels of development. First and foremost, this concerns governments, as part of their responsibility to regulate the interactions within society and the economy, manage risks and ensure fairness in society. The importance that the G7 and the G20 are giving to AI is a testimony of the urgency to agree on guidelines to steer AI development.
However, the speed of technological innovation and AI deployment remains a challenge, as the balance between innovation and regulation has often seen policies lagging behind, sometimes failing to reap the benefits of the changes. Only through a holistic approach, including all the relevant actors, can we respond to the transformations being driven by AI systems and prepare our societies, and our workers, for the transition to a future that is, to a great extent, already happening in the present.
A number of initiatives are trying to provide answers to some of these urgent questions. The tech community has been brainstorming on guidance for developers. The EU has produced a set of Ethics Guidelines for trustworthy AI. And now the OECD, as part of its mandate to deliver on better policies for better lives, has developed a set of AI Principles that will be adopted by 42 countries at the Organisation’s annual Ministerial Meeting on May 22, representing the first fully international public policy standards on AI. The OECD’s AI Principles take a broader approach than just ethics, and aim to help governments create national strategies to benefit and empower as many people as possible. The Principles, developed in consultation with a group of think-tanks, business, civil society, labour associations and other international organisations, focus on leveraging the benefits of AI while protecting human rights and democratic values, and ensuring fairness. They can also be used as a starting point from which to develop guidelines for use in fields such as science or genomics.
Ultimately, a set of collectively-agreed rules of the game will reduce complexity, foster innovation and exchanges on a global scale and help bridge the divide between technology and regulation. They are, however, just the beginning of the process to address, together, the big questions about the impact of AI in our lives.
- How can we further foster global co-operation, including developing countries and a diverse set of stakeholders, to encourage adherence to principles for trustworthy AI?
- How can we foster an agile but also controlled policy environment to foster the deployment of trustworthy AI?
- How can we effectively transpose AI international principles into national legal frameworks?
Continue the conversation and help us co-create the agenda
|OECD Forum 2019|
Find out more about OECD Forum 2019: World in EMotion