JamzNG Latest News, Gist, Entertainment in Nigeria
A court in Quebec has issued new guidelines limiting the use of generative artificial intelligence by judges, and stressing that judgment should essentially remain the responsibility of humans.
The guidelines, jointly adopted by the Quebec Court of Appeal, Court of Appeal, Quebec Court and municipal courts in the province, state that generative AI can be used for certain ancillary tasks but cannot replace legal reasoning, evidentiary assessment or judgment.
Judging, they say, should remain the exclusive responsibility of judges.
“Generative AI, no matter how sophisticated, has no judgment or conscience and is not subject to accountability of any kind,” the 10-page document said. “He is unable to understand the specific humanitarian, social and legal context of each case.”
The move comes months after questions emerged about whether Quebec judges might use generative AI in preparing decisions.
La Presse reported in March that the decision contained references to non-existent legal cases, raising the possibility that generative AI was being used. These events help create a sense of urgency within the judiciary as they work to establish rules for the technology.
Get daily National news
Get Canada’s daily news stories delivered to your inbox so you never miss the day’s top stories.
The court said there are currently no approved and safe institutional generative AI tools available to judges. They also warn that existing tools rely on algorithms whose workings remain largely unclear.
“Assessment cannot be reduced to a technical exercise,” the guidelines say. “This is an intrinsic human function, based on judgment, consideration of context, respect for dignity, requirements of justice and the exercise of moral responsibility.”
The court acknowledged that generative AI could be useful, but stated that it cannot be used as a source of legal authority or reasoning, to assess evidence or to make decisions.
Judges can use them for limited tasks such as proofreading, revising, redrafting or translating texts they have written, certain administrative work, or searching for information in certain sources.
Any use must remain under human control, and AI-generated material must be independently verified.
These guidelines make clear that judges remain fully responsible for their legal reasoning, the content of their decisions, and the accuracy of the facts, references, quotations and legal authorities they cite.
“There was no error that could be attributed to the tools used,” the document said. “The use of generative AI tools does not diminish the judge’s obligation to verify or their full responsibility for the final outcome.”
The guidance states that AI cannot be used for substantive legal reasons, analyzing or characterizing facts, assessing evidence or credibility, determining the outcome of a dispute, or drafting parts of a decision.
“If the use of AI has the impact or purpose of determining legal outcomes, assessing the credibility or interpreting disputed facts, interpreting the relevance of facts, or formulating reasons for a decision, then it should not be used,” the guidelines state.
The court also warned judges about the risks of confidentiality.
Judges may not provide AI tools with draft decisions, notes about ongoing cases, or information protected by law or court order.
Until a formal institutional system is in place, judges who use AI tools for permitted purposes should also ensure their settings prevent the information they enter from being used to train or improve the system.
When in doubt, the guidelines recommend judges not use generative AI.
These guidelines are intended to protect principles including judicial independence, impartiality, fairness, human rights and access to justice, while ensuring confidentiality, security and data protection.
© 2026 Canadian Press
