How Québec became a world-class AI powerhouse

The use artificial intelligence (AI) is exploding across the planet as it evolves into an essential tool for a myriad of fields and industries. But to successfully implement the technology into business operations, IT leaders need to surround themselves with global experts while building a state-of-the-art ecosystem.

The place to start looking for such experts in Canada is Québec, the nation’s AI powerhouse. Seventh in the world – that’s where the province ranks in the Global AI Index published by the British firm Tortoise Media, a ranking of the most competitive countries in AI.

[ Lisez la version française : « Comment le Québec est devenu une puissance mondiale en IA | CIO » ]

The 2021 Tortoise study ranks Québec just behind Germany and South Korea and ahead of technological powers such as Israel, Japan, and France. Canada overall comes in fourth place – a remarkable achievement of which Québec is a real driving force.

In December 2021, the Canadian supercluster Scale AI announced that three of its four new research chairs in artificial intelligence would be located in Montréal: AI for urban mobility and logistics at HEC Montréal, data science for retail at McGill University, and data-driven supply chains at Polytechnique Montréal.

How did the province achieve this meteoric rise? According to Tortoise, it’s a combination of three factors: the strength of its basic research, in which Québec ranks fifth among the approximately 60 countries reviewed; successful government investment strategies (sixth); and the emergence of a critical mass of companies (seventh).

According to Marie-Paule Jeansonne, president and CEO of Forum IA Québec, the province’s ecosystem now includes more than 550 AI companies. There are more than 25 incubators, accelerators, and venture capital funds, as well as 45 support organizations.

Even the global heavyweights in the field – DeepMind, Google, Facebook, Microsoft, Samsung, and Thales – couldn’t resist the Québec recipe and have all opened AI centres in Montréal.

From academia to business

Once a quiet field in academia, AI has been developing rapidly in Québec’s computer science departments over the past decade. McGill University and the University of Montréal currently have more than 300 researchers and PhD students in artificial intelligence – the largest concentration of academic AI talent in the world.

Among these international leaders, Yoshua Bengio is perhaps the best known. A pioneer in deep learning (DL) – which involves simulating how the brain works by building artificial neural networks – the University of Montréal professor won the 2018 Turing Award. He heads the Institut québécois d’intelligence artificielle (known as Mila from its former name, the Montréal Institute for Learning Algorithms), which alone brings together more than 900 deep-learning specialists.

Yoshua Bengio

It’s this more advanced variant of artificial intelligence that helped the technology take off at the end of the last century – so much so that AI and DL are now often used interchangeably.

Working in one of the world’s deep learning capitals, many Québec companies have spun off into fields using this technology: natural language recognition and understanding, medical image analysis for cancer detection, discovery of new biomarkers for various diseases, intelligent management of large buildings, digital data center infrastructure, aerospace, pilot training, and many other cutting-edge industries.

Philippe Beaudoin, president of Waverly AI, a startup that leverages computational language understanding, believes Bengio has played a crucial role in the emergence of the Québec ecosystem: “It’s obvious – especially since he’s the only one of the world’s deep-learning stars not directly linked to Big Tech. It’s this presence of a free agent that has allowed Montréal to attract the attention of companies and investors.”

“He also promoted the humanist and environmentalist values that ignite today’s bright young AI entrepreneurs.”

Artificial intelligence to improve customer experience

IVADO, the Institut pour la valorisation des données (Institute for Data Valorization), also headed by Bengio, is a Québec-wide organization at the interface of academia and private enterprise. This organization has proven to be a powerful catalyst for the commercial integration of AI in Québec:

“IVADO brought together the best AI researchers from Montréal universities – which subsequently attracted a lot of talent and businesses – and other researchers who now study how humans use new technologies,” said Sylvain Senécal, professor of marketing at HEC Montréal and co-director of the Tech3Lab, the world’s largest university laboratory for user-experience research.

The Tech3Lab is one of over 100 AI research centers in Québec. One of its major projects focusses on using artificial intelligence to understand the nature and intensity of emotions experienced by IT system users.

Pierre-Majorique Léger, a professor in the Department of Information Technology at HEC Montréal, is co-director of the lab with Senécal: “People generate a gigabyte of data per hour when they interact with IT systems,” he explained at the TimeWorld Global Congress on Artificial Intelligence in May 2022 in Montréal. “That’s the equivalent, in one hour, of the entire Encyclopedia Britannica.”

AI is essential to the management of such a multitude of data, and the effort can pay off handsomely. According to Léger, taking user experience into account at the design stage of a new product, rather than tailoring it down the line, can “reduce business costs a hundredfold.”

Philippe Beaudoin

A winning funding strategy

Government investments have also helped to propel Québec’s AI to the top. According to PwC Canada, the money invested was more than $800 million between 2017 and 2021, including $500 million from the federal government. Half of this helped fund various research centers, employing more than 3,000 skilled workers.

The injection of public funds has had a significant leverage effect on the integration of AI in many companies: the private sector raised more than $1.5 billion in venture capital for startups during this time.

“The federal and provincial strategies have been extremely well coordinated,” said Léger, whose lab is associated with both IVADO and the Canada Foundation for Innovation. In particular, he points out that it’s easier in Québec than in the U.S. to fund research projects along with the infrastructure they need – a significant competitive advantage.

A generation of innovative companies

Element AI, co-founded by Bengio and Beaudoin, also fostered a whole new generation of AI professionals. “We’ve attracted a lot of very good international talent that wasn’t academic enough for university labs,” Beaudoin said. “As a result, we allowed many people to pivot their careers to incorporate more AI and train in the field – facing real clients and real-world problems that they could address with AI.”

“We’ve really come into a second phase of our ecosystem development with this slew of new companies,” said Jeansonne. These include startups developing AI solutions ranging from social media to supply chains to homecare services.

“All this makes me proud as a businessman,” said Alain Lavoie, president of LexRock AI, a company that develops natural language processing tools for document management.

Lavoie believes that Québec’s success is based on the virtuous circle of basic research/technology transfer/commercialization. “What’s special about the world of AI is that it goes very fast. A company can’t keep up on its own,” said Lavoie.

To do it, LexRock AI also works with academics; it’s an extremely fruitful collaboration, Lavoie explained, although the partners’ needs and timelines are not always the same: “You have to know how to match the entrepreneur’s sprint with the researcher’s marathon.”

How sustainable is Québec AI?

Does Québec have all the tools it needs to stay at the forefront – or go even further? Jeansonne is cautiously optimistic: “There’s still a lot to be done. The next phase is the integration of AI [in the entire economy] to bring its advantages to all businesses.”

According to Lavoie, IVADO will continue to play a crucial role: “This organization does a lot for technology transfer,” he said – helping IT leaders define their projects, manage their expectations, and secure grants from organizations dedicated to building private-public partnerships – such as Mitacs, a national research organization linked to industrial and social innovation.

Forum IA Québec

In Québec, another promising way forward is to specifically link the AI ecosystem with new areas where local companies are making their mark. Announced in the spring of 2022, the E-AI initiative aims to integrate artificial intelligence into the entertainment industry; it already includes video-game publisher Behaviour Interactive and The Québec Film and Television Council – both in Montréal.

Beaudoin is counting on the expertise of Québec leaders to propel the next phase of development: “I think in the years to come, we will realize the quality of our ecosystem. Those who played a role in the first wave will be poking their noses everywhere: in large companies, in new start-ups – and as the new global leaders.”

Provided we maintain the passion for innovation, added Beaudoin: “There’s a risk of becoming too conservative in our future endeavors. It would be unfortunate to lose the huge advantage we’ve managed to create.”

Translation by Daniel Pérusse