Ottawa needs to collect more information about how artificial intelligence is impacting the Canadian workforce, a recent Parliamentary committee report says.
The report, published last month by a House of Commons committee on human resources, skills and social development, recommends Ottawa improve its understanding of how AI is impacting Canadian workers. Protecting workers’ privacy and addressing biases that AI tools may have about race and gender were key themes of the 44-page report.
But some experts who study AI say it is not clear that artificial intelligence is materially affecting Canadian jobs.
“For the most part, most Canadians don’t work in a workplace that’s exposed to AI and most people haven’t had to deal with it,” said Angus Lockhart, a senior policy analyst at the Dais, a think tank at Toronto Metropolitan University.
“[V]ery few people have been laid off because of artificial intelligence.”
‘Not felt yet’
The report recommends the federal government invest in skills training to help workplaces adapt to using AI.
It suggests Ottawa work with provincial governments and labour unions to create a framework to support workplaces in using AI in ethical ways. It did not say what the framework should include.
It also recommends that Statistics Canada create a way to monitor how AI is impacting the labour market in Canada, including examining whether particular jobs are at-risk.
The government has yet to respond to the report.
Lockhart authored a report, released last fall, that found only 3.7 per cent of Canadian businesses were using AI as part of their work in 2021. Despite the advance of AI tools since then, Lockhart still believes AI’s impact on people’s jobs and employment “has not really been felt yet.”
Typically, larger businesses are more likely to use AI more, says Laurent Carbonneau, director of policy and research at the Council of Canadian Innovators, an organization that advocates for innovation policy and supports Canadian companies in scaling up.
There are very few large, Canadian-started companies, says Carbonneau. Most Canadian businesses are smaller, which means they often do not have the resources to study how to use AI in their work.
“When you do not have big, homegrown companies, that tends to kind of slow down your technology adoption in your broader economy generally,” he said.
“You are not generating as much homegrown managerial talent that then goes and starts their own businesses or gets hired as an executive to a medium-sized company and helps them run themselves more like a big company.”
The Parliamentary committee recommends that the government specifically support small businesses and nonprofits across Canada as they increase their use of AI.
Privacy concerns
One of the main concerns identified in the report was how AI in the workforce may negatively impact workers’ privacy. The committee recommended the Office of the Privacy Commissioner of Canada review how AI is impacting workers’ privacy and create regulations to ensure Canadians are protected.
The committee heard multiple views on the topic. While some people who spoke at committee meetings said AI puts workers’ privacy at risk, others said current privacy laws are strong enough to respond to potential threats from AI.
“Digital governance is really hard, because things change so quickly,” Carbonneau said. In his view, current privacy laws already address privacy concerns well.
The government needs to consider using governance tools — such as voluntary industry standards — that are more flexible and less time-consuming than writing new laws or reviewing existing ones, he says.
Voluntary standards already exist in a number of industries in Canada, including standards about digital privacy. Governments would need to enact them into regulations to make them mandatory.
‘A fine line’
The report repeats concerns about Canada’s lagging productivity, a topic Canadian Affairs has reported on previously. The committee recommends the federal government explore using AI to help reduce government bureaucracy and red tape.
AI can increase productivity in a lot of ways, says Lockhart.
When most people think about AI, they think about things like ChatGPT. But ChatGPT is, in his opinion, “a small step instead of a massive leap,” said Lockhart.
AI best increases productivity by automating administrative tasks so workers can spend more time on other tasks, he says. When AI helps increase productivity, this means that overall quality of life for Canadians will improve, he says.
But balancing concerns about AI’s effect on privacy and its potential to increase productivity is difficult.
“There’s a fine line to walk here, between regulating AI in such a way that keeps people safe, without necessarily putting so much red tape in place that you don’t get any benefits from it,” said Lockhart.
In a dissenting opinion on the report, Conservative members of the committee said the government needs to take privacy more seriously and adopt legislation that will address concerns about AI.
“Artificial intelligence will soon be present in a plethora of workplaces across the country and ensuring the protection of workers’ private information in a variety of settings is an issue the government will quickly face,” the Conservative committee members wrote.
“Developing policy for all workplace contexts is essential to ensuring the protection of workers’ rights.”
