Ontario Watchdogs Release Urgent AI Guidelines Amid Regulatory Delays

Two of Ontario’s top oversight bodies have stepped in to fill a regulatory gap on artificial intelligence, releasing a new set of principles aimed at guiding the responsible use of AI across the province’s public and private sectors.

Ontario’s Information and Privacy Commissioner (IPC) and the Ontario Human Rights Commission (OHRC) unveiled the joint document this week, describing it as an urgent response to the rapid expansion of AI technologies and the Ford government’s still-unfinished efforts to create a formal regulatory framework.

“AI is evolving very rapidly,” said Patricia Kosseim, Ontario’s Information and Privacy Commissioner. “The deployment and development of AI in the public sector across Ontario is of great interest and priority for many institutions. We felt it was urgent to remind institutions of their existing legal obligations.”

The guidance is intended to help governments, public institutions, and private organizations decide how and when to deploy AI systems and, just as importantly, when those systems should be paused or shut down altogether.

Kosseim said her office has already received several complaints related to the growing use of AI in Ontario, prompting investigations into how the technology is being used and whether it complies with privacy laws.

In one case, students at a university raised concerns over AI-powered online proctoring software that monitored them during exams. The complaint led to an investigation and new guidance on how institutions can use such tools responsibly, balancing academic integrity with students’ privacy rights and ensuring the information collected is accurate and proportionate.

Human rights officials have raised similar red flags, particularly around bias embedded in AI systems.

“Without guardrails, biases in AI-driven research and decision-making can lead to unintended consequences,” said Patricia DeGuire, Ontario’s Chief Human Rights Commissioner. “These impacts often fall hardest on historically marginalized individuals and groups.”

The new document lays out a series of principles that should govern AI use in both public and private settings. Among them: AI systems must produce valid and reliable information; their use must be transparent and accountable; and their application must respect and affirm human rights.

Organizations are urged to conduct rigorous validity and reliability assessments before deploying AI tools and to reassess them regularly to ensure accuracy over time. The guidance also stresses the need for strong security safeguards to prevent unauthorized access or misuse of personal information.

A particularly notable warning concerns civil liberties. The commissioners caution that AI systems must not be used to “unduly target” individuals involved in public protests or social movements, or otherwise infringe on Charter-protected rights such as freedom of expression and assembly.

Perhaps the strongest language in the report relates to risk management. The commissioners say AI systems should be “temporarily or permanently turned off or decommissioned” if they are found to be unsafe or harmful. Institutions should also have clear processes in place to review and respond to negative impacts on individuals or communities.

“An ounce of prevention is better than a pound of cure,” DeGuire said. “We want the people of Ontario to benefit from AI, but as social justice oversight bodies, we must take the lead in preparing institutions and citizens for responsible innovation.”

The release of the guidance comes as the Ford government continues work on AI regulations first enabled by legislation passed in 2024. The Enhancing the Digital Security and Trust Act granted the province authority to regulate AI use in the public sector, emphasizing responsible, transparent, and secure deployment.

At present, however, only high-level principles apply directly to provincial ministries. Detailed regulations which would extend binding rules to Crown agencies, hospitals, schools, and other public institutions have yet to be finalized.

“When those regulations are eventually adopted, we hope sooner rather than later, they will provide binding parameters for all public institutions across the province,” Kosseim said.

While the guidance is focused on public institutions, the OHRC said it is also meant to set expectations for the private sector. Employers in Ontario are already required to inform job applicants when AI is used in hiring decisions, but the commission has warned that algorithmic bias poses a growing risk of indirect discrimination.

“We’re looking for broader safeguards in the use of AI in employment,” DeGuire said.

Ultimately, both commissioners said the goal is to ensure that AI strengthens public services and innovation without eroding trust or rights.

“Responsible use of this technology is essential,” Kosseim said, “so that AI benefits individuals and communities rather than undermining public confidence.”