Report calls on Quebec to act quickly to regulate AI

FILE - The OpenAI logo is seen displayed on a cell phone with an image on a computer monitor generated by ChatGPT's Dall-E text-to-image model, Dec. 8, 2023, in Boston. (AP Photo/Michael Dwyer, File)

Quebec needs to start working now on legislation to oversee the development and deployment of artificial intelligence (AI) in the province, according to a Quebec Innovation Council report unveiled on Monday.

The long-awaited report outlined the directions Quebec must take to maintain its leadership position in this field while protecting citizens with guidelines to ensure that the use of AI is ethical and responsible.

Quebec Economy Minister Pierre Fitzgibbon, MNA responsible for digital issues Éric Caire and Quebec's Chief Innovator, Luc Sirois, presented the 142-page report entitled "Ready for AI."

"Being ready for AI is more than a framework, it's a proactive posture on the part of a nation. It expresses a cautious but confident spirit, even a little rebellious, in the face of this new technology and all its implications, threats and promises," said Sirois.

For his part, Fitzgibbon revealed his government sees AI first and foremost as a formidable tool for economic development while acknowledging that it does present risks.

"For the government, it's imperative that AI be regulated. We want a framework, but one that doesn't put the brakes on innovation. Innovation is synonymous with the creation of socio-economic wealth. What I want is for AI to be anchored to our needs and endowed with trustworthy governance to reassure businesses and the public," he said.

The report's first recommendation is that "Quebec should immediately begin the work that will lead to the adoption of a framework law specifically dedicated to overseeing the responsible development and deployment of AI in society."

It further recommends removing the responsibility for implementing this framework law from the government and entrusting it, as well as the development of regulations related to its implementation to "an authority independent of the executive branch."

Work and culture

Since it is already a given that AI will generate major changes, the experts call on the government to revise and modernize labour law and social policies without delay to take account of the rapid evolution of AI.

"We want to prevent the rapid transformation of the world of work due to AI from accentuating social inequalities and slowing down the social and professional acceptability of AI deployment," said Sirois.

Caire spoke of the support needed for people whose jobs will be altered or eliminated by AI's introduction.

"We have to support people in this transformation of society. We need to give them the opportunity to either go and get additional qualifications or reorientate themselves towards other sectors of activity where we can support them and enable them to continue their careers," he said, acknowledging that this is "a concern."

On the socio-cultural side, they raise the need for action to counter misinformation and disinformation facilitated by AI's ability to produce deepfakes.

"On the side of democracy, AI amplifies misinformation. This calls on us to react proactively to protect this democracy, which is an essential foundation for society," argued Sirois.

Caire says he is well aware that many disinformation initiatives originate outside Quebec.

"That doesn't mean we can't do something," he said, referring to the various legislative initiatives being prepared around the world.

He also points out that "in Canada, forgery and the use of forgeries are already in the Criminal Code. It is a criminal act to deliberately seek to mislead someone," although it is virtually impossible to stop people in countries such as Russia or China.

The report also calls on the government to protect artists' work and copyrights, notably by preventing the use of their voice or image by generative AI without their consent.

At the same time, the Council encourages Quebec to use the legislative levers at its disposal to improve the visibility of francophone and Indigenous cultural content on major platforms such as Netflix and Spotify.

Training young people

Another of the report's major points is aimed at training, starting with the implementation of teaching programs from kindergarten to university in digital literacy and AI not only for use, but also to learn how to think critically about it.

Training must also be extended to various levels involved in professional and continuing education with workers to support their adaptation to AI's entry into the workplace.

For responsible implementation and development, the Innovation Council believes it is necessary "to invest significantly in research work, especially in the human and social sciences," among other things, to better identify the issues surrounding AI safety and its societal impacts.

It also suggests supporting the development and commercialization of the tools that will be developed with these technologies, and increasing the computing power to which players in the field have access, as AI requires very high-power computing capabilities.

Also, noting the "low quantity of French-language data and Quebec data," the report recommends supporting the development of a high-quality national cultural data bank.

Major investments

Finally, the report calls on the Legault government to make available immediately to ministries, agencies, Crown corporations and municipalities a large-scale budget envelope for planning and implementing projects in AI "strategic assessment," suggesting in passing that they focus on "the essential missions of the State, be they caring for the sick, educating children or combating climate change."

The report is the result of extensive consultation and reflection involving hundreds of experts and thousands of people from all spheres of civil society since last April.

Positive but cautious reactions

The labour movement, which took part in the consultations, says it cautiously welcomes the Council's report.

The Syndicat de la fonction publique du Québec (SFPQ) sees it as "a step forward in recognizing the challenges posed by AI: the principles of respect for government employees, valuing their professional judgment and the quality of public services."

However, its president, Christian Daigle, deplores "a certain opacity surrounding any project to integrate AI into public administration" and calls on the government to maintain an ongoing consultation process, underlining the risk of numerous problems linked to its implementation.

The CSN, for its part, welcomes the recommendations and is also asking to be involved in the next steps, in particular "the adoption, as soon as possible, of a framework law."

Its president, Caroline Senneville, points out that "workers are the first to be affected by this technological revolution."

In the public sector, she calls for the utmost vigilance in the education sector because of the "risks of plagiarism and deterioration in the pedagogical relationship between students and staff."

Another of the participants, the Ordre des ingénieurs du Québec, points out that engineering will be directly affected by these new technologies and promises to identify measures to "support the supervision and professional practice of its members in the use of technologies involving artificial intelligence."

-- This report by The Canadian Press was first published in French on Feb. 5, 2024.