Canadian CIO group publishes national AI standard
A Canadian body representing chief information officers from the public, private and voluntary sectors has set out national standards for the ethical design and use of artificial intelligence (AI), in what it claims is a world first.
Over the course of a year the CIO Strategy Council, a not-for-profit organisation, engaged over 100 experts and thought leaders from various sectors and industries to create what it says is the world’s first “consensus-based” standard on the responsible use of automated decision systems.
The newly-established council, which aims to be a forum where Canadian CIOs can collectively transform, shape and influence the Canadian information and technology ecosystem, received accreditation from the Standards Council of Canada in April. It is the only accredited body in the country that focuses exclusively on technology standards.
Ethical design and use of automated decision systems is the council’s first approved and published standard, and outlines minimum requirements for protecting human values and incorporating ethics in the procurement, deployment and use of AI systems. The standard is intended for public and private sector companies, government entities, and not-for-profit organisations, and is available free of charge.
Setting the standard
It is based on accepted AI ethics principles, such as those set out by the OECD. These recommend that AI systems be designed in a way which respects the rule of law, human rights, democratic values and diversity; that there should be transparency and responsible disclosure around AI systems to ensure that people understand AI-based outcomes and can challenge them; that potential risks should be continually assessed and managed; and that organisations and individuals developing, deploying or operating AI systems should be held accountable for their proper use.
According to the CIO Strategy Council, the standard – which covers risk management, ethical considerations, and the deployment, maintenance and monitoring of AI systems – “goes beyond a common set of aspirational principles and provides a framework and process that can be both measured and tested for conformity,” and aims to give buyers “confidence in the technologies that are providing information, recommendations, or making decisions using AI and machine learning”.
The standards cover, for example, how AI should be handled within organisations’ compliance, risk management, impact evaluation, appeal and escalation, and planning functions. “Having this framework will allow us to bring a sense of order and apply a common yardstick of integrity against our implementation approach and value-based outcomes,” said Philip Armstrong, executive vice president & global chief information officer of Great West Life Co, who is a council member.
Agreed plan
The council says it was able to accelerate the standards-setting process in Canada – which can take years – by adopting an “agile, open and consensus-based approach”.
Alex Benay, Canada’s former chief information officer – now chief client officer at MindBridge Ai, and co-chair of the CIO Strategy Council with former Blackberry co-CEO Jim Balsillie – told Global Government Forum that the process is unique because it is open to everyone, and because it closes the gap between policymakers and technical expertise.
“At the table you have government, you have banks, you have people from the energy sector, from insurance, you’ve got NGOs, you’ve got provinces, and soon we’ll have municipalities too – so you’ve got all the players,” he said. “It’s a fully cross-sectoral approach whereby everything is participatory in nature – and on top of that the standard-setting process is completely transparent.”
The hope, he said, is that organisations adhere to the standards from the start of the procurement process. “In theory, what I’d love to see is organisations and banks and governments of all stripes in Canada not buy a piece of technology if it does not abide by the standard. That’s the ultimate goal.”
“By using the standards you’re guaranteeing that you’re not buying technology that, for example, would make decisions without its users being aware of it,” he added. “There are a lot of companies out there offering algorithms that aren’t explainable. Only buying systems that use explainable algorithms would increase responsibility.”
The CIO Strategy Council is also working on setting standards on the third-party usage of personal data, to address the commercialisation of data, and on digital identity. These, Benay said, will “create a framework to put everyone on the same page and act as the first step toward regulation”.
About Mia Hunt
Mia is a journalist and editor with a background in covering commercial property, having been market reports and supplements editor at trade title Property Week and deputy editor of Shopping Centre magazine, now known as Retail Destination. She has also undertaken freelance work for several publications including the preview magazine of international trade show, MAPIC, and TES Global (formerly the Times Educational Supplement) and has produced a white paper on energy efficiency in business for E.ON. Between 2014 and 2016, she was a member of the Revo Customer Experience Committee and an ACE Awards judge. Mia graduated from Kingston University with a first-class degree in journalism and was part of the team that produced The River newspaper, which won Publication of the Year at the Guardian Student Media Awards in 2010.
Related Posts
