Written by Jeff Pang, BSc, JD
Borden, Ladner, Gervais, LLP
www.blg.com
As artificial intelligence, or “AI,” becomes increasingly prevalent, from dedicated AI assistants to AI-enabled household devices, it is also being used more frequently in healthcare settings. The practice of veterinary medicine is no exception. This article provides a brief summary of the current regulatory landscape and offers practical guidance to veterinarians looking to implement AI tools in their practices.
Artificial Intelligence
Although AI appeared to surge in public awareness in 2022 with the rise of ChatGPT, it is a branch of computer science with a long history of machines simulating human cognition, including learning, reasoning, problem-solving, perception, and creativity, by training on large datasets to identify patterns. For example, in both human and veterinary medicine, AI software can learn from databases of patient records that list symptoms alongside corresponding diagnoses. By examining these records, AI can detect patterns and suggest which illness or condition may correspond to certain symptoms. Accordingly, veterinarians should understand the AI tools they use by reviewing their design, the training data used in their development, and the tool’s outputs, in order to assess reliability and to identify and mitigate bias.
AI tools being used
Veterinarians today often use AI tools to assist with clerical tasks, such as keeping medical records up to date, transcribing conversations, and sending communications and reminders to clients. However, an increasing number of AI tools are also being made available to guide veterinarians toward making diagnoses, including AI assistants and AI-enabled stethoscopes and other tools developed based on the datasets described above. These tools can be valuable because they may improve access to veterinary care by streamlining appointments, but they are not without risk. AI-fueled hallucinations, which occur when generative AI models generate confident, convincing, yet inaccurate or fabricated medical information, diagnoses, or references, remain a significant concern across the AI tool landscape. Bias is also a concern, as AI tools, particularly chatbots, may be steered toward the tendencies or preferences of its users.
Regulatory landscape
In human medicine, AI tools are treated as “Software as a Medical Device” and require Health Canada approval for use in clinical settings. Veterinarians should note, however, that veterinary platforms based on human systems are not always translatable to animals. At present, no regulatory process exists for AI tools used in veterinary practices, creating a regulatory gap in which veterinarians may incur personal liability if they are not discerning about how, and which, AI tools they use in their practices.
The Canadian Veterinary Medical Association (the “CVMA”) has issued a position statement urging that scientific rigor be applied to the development of AI tools through an effective national policy framework and recognized standards, in order to mitigate risk to patients and reduce liability exposure for veterinarians. The CVMA also states that AI tools should be delivered to veterinary professionals in accordance with the policies of the applicable provincial or territorial veterinary regulatory body and urges these bodies to develop policies and standards for AI applications used in veterinary medicine. The stated objective is to mitigate risk to professionals, clients, and patients by supporting accuracy, transparency, and good practices.
With respect to regulatory bodies, the College of Veterinarians of Ontario (the “CVO”) is the only provincial regulatory body that has issued substantive guidance on this topic. The CVO emphasizes that, as previously discussed, other than medical devices that use forms of energy (such as imaging and laser equipment), no medical device for use in animal health care is subject to oversight or approval by Health Canada. The CVO also stresses that once a veterinarian chooses to use AI, they remain responsible for responding appropriately and potentially carrying out the AI’s recommendations.
The CVO further advises that, when assessing an AI tool, veterinarians should ensure its purpose, function, and relevance are clearly defined and justified. Veterinarians should also evaluate the quality and source of the data underlying the tool, identify any inherent biases, and understand how those biases are addressed. In addition, veterinarians should weigh the potential risks and benefits the tool may bring to animal care and consider how it could support their professional competence. The data used by the tool should be accessible and manageable for veterinarians, with privacy and security safeguards and clear policies governing ownership and use. Finally, the device should be updated regularly and properly maintained to support effective and timely veterinary care.
Best practices
Veterinarians should keep in mind that the use of AI tools does not absolve them of their professional responsibilities to their respective governing bodies, to the profession, or to the public. Their reputations and licenses remain at stake when employing AI, which should function as an aid rather than a replacement for professional judgment. AI tools commonly disclaim liability and emphasize the need for professional oversight. Accordingly, veterinarians retain full responsibility for decisions made in their practices. This principle will likely persist even if a regulatory framework is introduced, underscoring the importance of careful and discerning use of AI in veterinary medicine. Veterinarians should always disclose their use of AI tools to their clients as a matter of obtaining clients’ informed consent to do so. Veterinarians may also wish to develop acceptable AI-use policies for their practices. As a fundamental tenet of professional responsibility, veterinarians are responsible for the actions of their auxiliaries, and the use of AI tools should be approached in an informed manner.
If you have any questions or would like further guidance on implementing AI tools in your veterinary practice, please feel free to reach out to us. We would be pleased to support your efforts to help ensure the safe, effective, and responsible use of technology in veterinary care.











