Back
Canada's AIDA and Health Care AI Governance: Lessons from a Failed Regulatory Framework
webai.nejm.org·ai.nejm.org/doi/full/10.1056/AIpc2500153
Published in NEJM AI (June 2025), this paywalled article is relevant to AI governance researchers studying sector-specific regulation, particularly how general-purpose AI laws may inadequately address health care AI risks.
Metadata
Importance: 52/100journal articleanalysis
Summary
This NEJM AI article analyzes Canada's failed Artificial Intelligence and Data Act (AIDA), which was terminated with Parliament's prorogation in January 2025. It critiques AIDA's lack of specificity, underinclusiveness, and absence of sector-specific health care oversight, and proposes reforms for future AI legislation. The Canadian experience offers broader lessons for global AI regulation balancing innovation with patient safety.
Key Points
- •Canada's AIDA was terminated in January 2025 before enactment, leaving a regulatory gap for health care AI governance.
- •AIDA was criticized for being too broad, lacking sector-specific provisions for safety, bias, transparency, and patient privacy.
- •The article proposes targeted, sector-specific regulatory approaches as essential for safe AI integration in health care.
- •Canadian regulatory failure provides instructive lessons for global AI governance efforts, especially in high-stakes domains.
- •Balancing responsible innovation with patient safety requires more granular legislative frameworks than AIDA provided.
Cited by 1 page
| Page | Type | Quality |
|---|---|---|
| Artificial Intelligence and Data Act (AIDA) | Policy | 46.0 |
Cached Content Preview
HTTP 200Fetched Mar 20, 20268 KB
## This article is available to subscribers. [Subscribe now.](https://ai.nejm.org/action/clickThrough?id=128860&url=%2Faction%2FstoreProxy%3Faction%3Dsubscribe%26product%3Dnejmai%26url%3Dhttps%253A%252F%252Fai.nejm.org%252Fdoi%252Ffull%252F10.1056%252FAIpc2500153%26promo%3DOAIGTBAN&loc=%2Fdoi%2Ffull%2F10.1056%2FAIpc2500153&pubId=41400171&placeholderId=101349&productId=101633 "Subscribe Now")
[Subscribe Now](https://ai.nejm.org/action/clickThrough?id=128860&url=%2Faction%2FstoreProxy%3Faction%3Dsubscribe%26product%3Dnejmai%26url%3Dhttps%253A%252F%252Fai.nejm.org%252Fdoi%252Ffull%252F10.1056%252FAIpc2500153%26promo%3DOAIGTBAN&loc=%2Fdoi%2Ffull%2F10.1056%2FAIpc2500153&pubId=41400171&placeholderId=101349&productId=101633 "Subscribe Now")
Already a subscriber? [Sign in.](https://ai.nejm.org/action/clickThrough?id=128860&url=%2Fsign-in%3Furi%3Dhttps%253A%252F%252Fai.nejm.org%252Fdoi%252Ffull%252F10.1056%252FAIpc2500153&loc=%2Fdoi%2Ffull%2F10.1056%2FAIpc2500153&pubId=41400171&placeholderId=101349&productId=101633)
Save
Contents
## Abstract
Canada’s initial attempt at AI governance, the Artificial Intelligence and Data Act (AIDA), was introduced within Bill C-27, but was ultimately terminated with the prorogation of Parliament in January 2025. AIDA sought to establish a risk-based regulatory framework; however, it was criticized for its lack of specificity, underinclusiveness, and absence of sector-specific oversight — issues that are particularly consequential for health care AI applications. The broad and generalized nature of AIDA left regulatory gaps concerning safety, bias, transparency, and patient privacy in AI-driven medical decision-making. In this article, we analyze the shortcomings of AIDA in health care AI governance and propose key reforms to guide future legislative efforts. A targeted, sector-specific approach is essential to ensure AI’s safe and effective integration into health care while fostering responsible innovation. The Canadian experience provides important lessons for global AI regulation, particularly in balancing technological progress with patient safety and ethical considerations.
Already have an account? [Sign In](https://ai.nejm.org/action/clickThrough?id=128880&url=%2Fsign-in%3Furi%3Dhttps%253A%252F%252Fai.nejm.org%252Fdoi%252Ffull%252F10.1056%252FAIpc2500153&loc=%2Fdoi%2Ffull%2F10.1056%2FAIpc2500153&pubId=41400171&placeholderId=101345&productId=101633)
Continue reading this article. SELECT AN OPTION BELOW:
### This Article is Available to Subscribers
[Subscribe Now for Full Access](https://ai.nejm.org/action/clickThrough?id=128880&url=%2Faction%2FstoreProxy%3Faction%3Dsubscribe%26product%3Dnejmai%26url%3Dhttps%253A%252F%252Fai.nejm.org%252Fdoi%252Ffull%252F10.1056%252FAIpc2500153%26promo%3DOAIGBAAN&loc=%2Fdoi%2Ffull%2F10.1056%2FAIpc2500153&pubId=41400171&placeholderId=101345&productId=101633)
Are you a member of an institution such as a university or hospital? [Learn more about Institutional Access](https://ai
... (truncated, 8 KB total)Resource ID:
42365d7c4104a03d | Stable ID: MTcwMjEwZj