Amazon hopes Canadian AI charter will enable interoperability

(Toronto) Amazon Web Services’ director of artificial intelligence encourages Canada not to go it alone when it comes to regulating technology.

Canada should settle for AI legislation that is interoperable with the safeguards that other countries will eventually use, otherwise many booming companies could find themselves in trouble, Nicole Foster warned Tuesday.

“Many of our young companies are wonderfully ambitious and hope to be able to sell and do business around the world, but having rules that are tailored and unique to Canada is going to be an extremely limiting factor,” said Ms.^me Foster speaking at the Elevate tech conference in Toronto.

Canada is working on an artificial intelligence and data law that aims to define how the country will design, develop and deploy the technology.

The legislation is still working its way through the House of Commons and is not expected to come into force until at least next year, but it is being closely watched in the tech sector and beyond.

Many fear the legislation will stifle innovation and cause companies to flee to other, more AI-friendly countries, but most agree the tech industry cannot be left to decide its own matters. -crazy.

They believe the industry needs certain parameters to protect people from systems that perpetuate bias, spread misinformation, and cause violence or harm.

As the European Union, Canada, the United States and several other countries each chart their own path toward regulations, some in the tech community have called for collaboration.

M^me Foster says there are “really promising signs” that this could come to fruition based on what she has seen in the G7 countries.

“Everyone says the right things. Everyone thinks interoperability is important, she said, but saying it’s important and doing it are two different things. »

Canadian Industry Minister François-Philippe Champagne is largely responsible for the approach the country will take to AI.

Last summer, he told attendees at another tech conference in Toronto that he believed Canada was “ahead of the curve” on artificial intelligence, even surpassing the European Union.

His government said it would ban the “reckless and malicious” use of AI, establish oversight by a commissioner and the industry minister who would impose financial sanctions.

Whatever Canada’s decision, M^me Foster warned that he must be “mindful of the cost of regulation” because requiring companies to undergo assessments to ensure their software is secure can often take time and much of this work is already in progress.

She believes the best regulatory model will identify high-risk AI systems and ensure measures are put in place to mitigate the harm they could cause, but not regulate things that shouldn’t be regulated .

Among the AI ​​systems that she says can do without regulation are mundane systems, like those that help deliver luggage to travelers at the airport more quickly.

“I think we need to focus on the risks that we need to manage and not hinder the use of really valuable technologies that will improve our lives,” argued M^me Foster.

The role of businesses

During another panel, Emily McReynolds, head of global AI strategy at Adobe, also mentioned that businesses also have a role to play in the regulatory debate.

Adobe, she said, has committed not to mine the web for the data it uses in its AI systems and has instead opted to obtain information licenses. She presented the move as a move that brings transparency to the company’s work, but also ensures that it “truly respects creators,” who tend to use the company’s software.

Adobe has chosen to take a proactive approach on issues such as data and has told other companies “it’s really important to understand that creating responsible AI is not something that comes after.”