A panel at the fourth edition of All In 2026 in Montreal debated whether AI can help address Canada’s shortage of primary care providers. The discussion took place Thursday at the Palais des congrès de Montréal during the major artificial intelligence conference.
Approximately 5.9 million Canadian adults lack a designated primary care provider — such as a family doctor or nurse practitioner — according to the 2025 “NosSoins” survey conducted by Unity Health Toronto in partnership with the Canadian Medical Association.
Muhammad Mamdani, director of the Temerty Faculty’s Centre for AI Research and Education in Medicine and Clinical Lead for AI at the University of Toronto, told the panel that AI could assist patients who do not have a primary care provider. To illustrate, he described a patient at home feeling unwell who does not know what is wrong and may seek care in an emergency department if they lack a family doctor. “They analyze this information mentally to try to make a diagnosis. Can AI do that?” Mamdani asked. He said the answer “requires nuance, but it leans toward the affirmative.”
AI tools and potential applications
Mamdani cited Google’s project, AMIE (Articulate Medical Intelligence Explorer), which enables video consultations with AI. “It will hear your cough. It will see you wrapped in a blanket. It will know you have the chills. So, you begin providing it with this multimodal data. We have six million people without a primary care physician. Can we now start to remedy this situation using AI? It won’t replace doctors, I hope. But it will certainly alleviate the pressure,” he argued.
In June, the federal government launched “AI for All,” Canada’s new national AI strategy. During Thursday’s panel, Elissa Strome, executive director of the Pan-Canadian Artificial Intelligence Strategy at the Canadian Institute for Advanced Research (CIFAR), highlighted the “significant” focus placed on health care. “About 30 per cent of our 150 CIFAR Canada AI research chairs are working to advance the science and application of AI within the health-care system,” she said.
The research institute has several long-term projects in the pipeline, including an AI application “that integrates medical imaging across a range of different modalities” and an application that “is developing a dashboard for diabetes prevention and prediction.”
Public trust, policy and infrastructure
Public trust in artificial intelligence and data protection were cited repeatedly by panel experts as major challenges. Strome stressed the need for broad engagement: “We need to have a dialogue with the public to better understand their fears and highlight the benefits. We need to include patients, clinicians, and developers in a single conversation.”
The new “AI for All” strategy also plans for the development of a world-class AI supercomputer and investment in sovereign computing power and cloud infrastructure. The government aims to strengthen measures protecting Canadians’ personal information, including against deepfakes, and to establish an online safety framework to better protect users of social media and chatbots.
