Why Canada needs to take the plunge on AI

Canada has received acclaim around the world for supporting researchers, like Dr. Jeffery Hinton, who pioneered the deep learning approaches that have led to an explosion in artificial intelligence (AI) technologies. Accenture estimates AI has the potential to double economic growth rates by 2035, and in the private sector, we are witnessing tremendous adoption of AI to increase productivity and competitiveness. In the public sector, as is too often the case, adoption is lagging significantly behind. Machine learning and AI have huge potential to make human decision-making in government vastly more intelligent and efficient—and to automate old, manual ways of doing things. Too often when you look beneath the hood of a provincial ministry or federal department, they rely on manual processes and spreadsheets to complete processes that in the private sector would have been digitized and automated years ago. This isn’t just about making government less expensive, but designing services better, so they respond to the needs of Canadians. It’s also about using technology to move past the bureaucracy and internal bottlenecks that regularly delay service delivery. Around the world, many governments are charging forward with AI to make government work better. For example, the U.S. National Weather Service is using massive amounts of data and machine learning to more accurately predict hurricane pathways. Many jurisdictions are using chat bots and virtual assistants to help citizens—especially the elderly or less technology inclined—to quickly get answers or services from government. Beyond what has been accomplished in these relatively early years of AI in government, the potential future use cases are immense. As severe weather events increase, AI could identify impacted communities and automatically ensure response organizations are co-ordinated and relief resources are in place when and where they are needed. AI can analyze patient interactions with provincial health systems at a massive scale to identify patterns and treatment best practices to make the optimal use of system resources and deliver better care. To deliver on the immense potential of AI in government, there are three core challenges governments need to overcome. First, the public is becoming increasingly conscious about large-scale collection and use of data whether by governments or private corporations, and rightly so. There are legitimate concerns that artificial intelligence could reinforce systemic bias and corrode personal privacy. For governments, and as a global society we need to develop a social consensus around the ethical use of AI that delivers the benefits of this technology, while protecting individual rights and freedoms. This is an area where Canada has boldly been leading the way. The government of Canada has established guiding principles for the ethical use of AI in government, and an Algorithmic Impact Assessment process that helps ensure bias or inequality is rooted out early. Canada has also established an Advisory Council on Artificial Intelligence, focused on trust and “human-centric” approaches to AI, and the Canada’s CIO Strategy Council has released standards for ethical design of automated decision systems. Many machine learning use cases depend on access to large amounts of data. The second major challenge facing government is that for too long the siloed nature of bureaucracies has led to data being isolated within particular ministries and departments. Classifying more data sets as open data is part of the answer, and Canada is a leading member of the Open Government Partnership that is driving this global movement. But often, legislative frameworks prevent public organizations from sharing and comparing data internally that could deliver insights on how individuals interact with a range of government services, or how millions of patients navigate the health system. In many places in Canada this is also starting to change. The federal Privacy Act (which applies to internal government data collection) is currently being reviewed, and Ontario’s Ministry of Health has discussed amending its health privacy legislation to enable easier information sharing within the system. While protecting Canadian’s privacy needs to be paramount, removing legislative barriers will create openings for AI and automation to make the system work better for everyone. The third challenge governments need to overcome is cultural—how to make a cautious, hierarchical organization structure designed for the 19th century responsive to fast-changing technologies and automation. There is a need to break down barriers between policy, program delivery, and technology teams inside government and move to a place where data insights drive decisions, and technology solutions, including AI, are considered at the front end. Procurement needs to move from a focus on the lowest price to getting the greatest value. Canada’s federal government is demonstrating leadership in this space as well. Last year, it introduced an agile procurement vehicle for AI tools that lets departments experiment with technology partners to solve a problem, and not just buy a piece of software. Other governments in Canada need to take a similar approach and start looking at technology vendors as partners rather than suppliers. The Canadian public played a major role in funding the development of AI technologies that will make the world a better place a million times over in the next century. For their part, Canada’s governments need to embrace the AI revolution and make sure Canadians get the smart digital government they deserve. Bob Osborn is ServiceNow’s chief technology officer, federal public service. The Hill Times

Why Canada needs to take the plunge on AI

Machine learning and AI have huge potential to make human decision-making in government vastly more intelligent and efficient.

AI and mental health care: opportunities and oversight

The Artificial Intelligence and Data Act needs a reset

Liberals on collision course to entrench anti-Blackness

AI and mental health care: opportunities and oversight

Canada needs a modern, national health data architecture

To address online harms, Canada must commit to digital media literacy

A cure for multiple sclerosis could be just around the corner

Renewable energy is the way to go, says letter writer

Supporting local broadcasters is critical for our democracy

Medical professionals’ discomfort with digital tech is stumbling block to spread of AI in health care, say experts

To take AI to the next level, we need a dose of wisdom

AI innovation needs federal attention, say industry players, lawmakers

The robot revolution is changing jobs and businesses in Canada

Three ways AI can substantially improve economic analyses

News

Get free news updates

Opinion

Submit a guest opinion article

resources

Advertise

subscribe

get help

Why Canada needs to take the plunge on AI

Machine learning and AI have huge potential to make human decision-making in government vastly more intelligent and efficient.

Get The Policy Briefings Newsletter