Trudeau says powering AI without compromising climate change is a G7 priority

“To do so, we must build the infrastructure necessary to achieve this at the speed that matches AI development, and nuclear technologies will be a crucial part of the solution.”

The prime minister added that at the same time “just as much thought” needs to be put into ways to reduce the energy demand of AI. The technology notably requires a significant amount of electricity for necessary computing power.

At an event later in the day where he appeared with French President Emmanuel Macron, Trudeau said now is the moment to decide how to act on AI, including on “how we make sure that we have the energy to power AI in a clean enough way and an efficient enough way that’s not going to set us back in our fight to reduce… carbon emissions.”

Trudeau also spoke about ensuring countries around the world benefit equally from AI. He said the technology will have a similar effect as the development of electricity, but there are still parts of the world that lack electricity.

He said if there is a similar delay in adopting AI, the consequences will be even more catastrophic and painful, “not just for those that don’t have access to it, but for the divisions and conflicts that will result.”

The AI Action Summit is expected to include nearly 100 heads of state and government and almost 1,000 civil society stakeholders from about 100 countries. Experts say it offers a chance to demonstrate Canada’s strength when it comes to artificial intelligence.

Florian Martin-Bariteau, research chair in technology and society at the University of Ottawa, said “it’s important to position Canada and to remind people that Canada has something to say.”

Martin-Bariteau said Canada has been a leader in AI governance, citing the 2017 Montreal declaration on responsible AI. In 2020, Canada and France also jointly launched the Global Partnership on Artificial Intelligence, which is now being integrated with the OECD. Trudeau said Sunday Macron was on the same page as him early on when it came to AI, beginning conversations in 2018 at the G7 meeting at Charlevoix.

Both initiatives predate the emergence of widely-available generative AI that has invigorated the conversation around AI safety and regulation in the past few years.

Since the launch of ChatGPT brought the issue into the forefront, there have been two international meetings, largely focused on AI safety and risk. The first took place at Bletchley Park in the U.K. in 2023, and the second in Seoul, South Korea, in 2024.

The two-day summit in Paris will have a broader focus, looking at questions including AI and the public interest, the future of work, innovation and culture.

Rowan Wilkinson, a research analyst for the digital society program at Chatham House in the U.K., said Trudeau’s attendance demonstrates and supports current Canadian policy efforts and said the country’s “academic excellence on this topic also needs to be embraced.”

“The summit is an opportunity for Trudeau to signal his commitment to this transformative technology and Canada’s role on the global stage,” she said.

But while Canada’s research strength in AI has been widely recognized, critics have taken issue with some aspects of the country’s AI efforts, including that it has been slow to commercialize the technology.

And while the Liberal government has lauded its AI regulation bill, some said the government was too slow to get it through Parliament, and the bill now looks likely to die on the order paper ahead of an anticipated spring election.

Martin-Bariteau said the joint statement on AI safety issued at the first meeting in Bletchley Park in the United Kingdom was one of the successes of that 2023 summit, and most of the countries who signed on are now “building capacity at the national level.”

The countries represented at that meeting also asked Yoshua Bengio, one of the Canadian “godfathers” of AI, to lead efforts to put together an international report on AI safety. The report, which incorporates input from 96 AI experts from around the world, was released in late January and focused on general-purpose AI.

It says some of AI’s potential harms are well-known, including the use of technology for scams, non-consensual intimate images, child sexual abuse material, the risk of bias in system outputs and privacy violations.

It says as general-purpose AI gains more capabilities, more risks are emerging, including “large-scale labour market impacts, AI-enabled hacking or biological attacks, and society losing control over general-purpose AI.”

Wilkinson said Bengio’s report is “a timely, comprehensive reminder of the risks and capabilities of general-purpose AI systems and will likely be a conversation-starter over the summit.”

At the 2024 AI summit in Seoul, world leaders agreed to build a network of publicly backed safety institutes to advance research and testing of the technology. Canada announced the launch of the Canadian Artificial Intelligence Safety Institute in November.

Nicolas Papernot, co-director of the institute’s research program, said its work will be grounded in Bengio’s report as it pursues questions about bias, the robustness of AI systems and their predictions and the potential for privacy breaches.

Papernot said the Canadian institute is “in touch with a lot of these parallel institutes. The idea is essentially to have a shared agenda for research.”

This report by The Canadian Press was first published Feb. 9, 2024.

Anja Karadeglija, The Canadian Press

by Canadian Press

Trudeau says powering AI without compromising climate change is a G7 priority

This Week in Flyers

This Week in Flyers

Our Community