Ontario didn't wait for a dispute to decide this — it wrote a definition of artificial intelligence directly into its guide to the Employment Standards Act, specifically because employment decisions were starting to be made by systems.
Short answer
Yes. Nothing in the ESA carves out an exception for a decision a system generated rather than a manager. Ontario's guide to the Act defines “artificial intelligence” as “a machine-based system that, for explicit or implicit objectives, infers from the input it receives in order to generate outputs such as predictions, content, recommendations or decisions” precisely because the Act needed language for this scenario.
Ontario's ESA guide states in full: “Artificial intelligence (AI) — A machine-based system that, for explicit or implicit objectives, infers from the input it receives in order to generate outputs such as predictions, content, recommendations or decisions that can influence physical or virtual environments… Whether a particular system falls within the scope of the definition of ‘artificial intelligence’ under the ESA will depend on the specific facts of each case.” That definition currently anchors one specific duty — disclosing AI use in a job posting — but its existence answers the broader question: the legislature already contemplated a system doing the deciding.
The ESA's substantive floors — hours-of-work limits, termination notice, the rest of the Act — attach to the employer's conduct, not to who or what produced a given decision. Canada's privacy commissioners make the same point from the accountability side: organizations using a generative AI system should “know that accountability for decisions rests with the organization, and not with any kind of automated system used to support the decision-making process.” An ESA obligation doesn't loosen because a system, rather than a person, produced the outcome.
A hiring decision generated without a person reviewing it, and a shift schedule an optimization tool assembles, are both instances of the same underlying question — and each has its own specific answer under Ontario law.
See also: the hiring version of this question, the scheduling version of this question, why ordinary statutes, not a dedicated AI law, do this work generally.
Keeping an automated decision correct, accountable and compliant once it's running is operations work — ai-operations covers it.
