Ontario Proposes Data Centre Approval Framework

Ontario opened a 30-day consultation on August 13 for a proposed framework governing new data centre connections as AI infrastructure demand rises. The proposal would create a separate electricity rate class for new large facilities, generally those above 1 MW, and require projects to cover their electricity-related costs rather than shift them to existing ratepayers, according to the provincial consultation notice.
Ontario has opened a 30-day public consultation on a proposed framework for approving new data centre connections to its electricity grid, responding to a rise in prospective projects tied to AI, machine learning, and cloud computing. The Ministry of Energy and Mines' consultation notice, posted August 13, proposes a separate electricity rate class for new large data centres and an approval process for facilities seeking to connect or reconnect to the grid.
The consultation, ERO number 026-0853, remains open until September 12. Ontario has not released a final regulation or final version of the framework.
Connection approval and electricity costs
According to Ontario's Environmental Registry notice, the province is considering a regulation under the Electricity Act, 1998 that would require new large data centres to obtain government approval before connecting or reconnecting to the grid. Facilities that do not meet the eventual requirements would be unable to connect or reconnect, the notice states.
The Ministry of Energy and Mines is also proposing a new rate class for large new data centres, using projects greater than 1 MW as an example threshold. The stated purpose is to ensure sector growth does not raise electricity bills for existing customers.
The Globe and Mail reported that Premier Doug Ford described the proposed threshold as applying to essentially every newly built AI-capable data centre. Ford also said developers would need to fund necessary infrastructure upgrades, while the provincial government ruled out financial incentives for construction, according to the newspaper.
BNN Bloomberg reported that Ford said Ontario would not offer new data centres a "freebie" and described the facilities as part of the "way of the future." The provincial proposal, however, remains subject to consultation and a final government decision.
Three assessment pillars
BetaKit reported that the proposed playbook organizes approval assessments around three pillars: economic development, digital sovereignty, and community benefits. Ontario's consultation notice similarly states that the assessment would evaluate prospective facilities for their economic and strategic value to the province, communities, and Ontarians.
The Globe and Mail reported that the framework would also consider local economic benefits and environmental standards related to water consumption and noise. These issues have become prominent in municipal discussions around large AI data centre developments, which concentrate high-density computing equipment and can impose substantial power and cooling requirements.
CityNews reported that municipal councils would not receive a veto under the announced framework. At the same time, BNN Bloomberg reported that Oakville enacted a one-year data centre moratorium and that Mississauga was scheduled to consider its own pause. Reporting cited residents' concerns about noise, water, heat, air pollution, grid strain, and utility costs.
Why the proposal matters for AI infrastructure
Ontario's proposal places grid interconnection, rate design, and local-benefit criteria alongside conventional project economics. For operators building GPU-intensive training and inference capacity, the relevant constraint is not simply whether a site can obtain power, but whether it can satisfy the jurisdiction's connection and cost-allocation rules.
High-density AI data centres make electricity use and local requirements central to infrastructure planning. Ontario's consultation is therefore relevant beyond Canada as an example of policymakers tying AI infrastructure expansion to ratepayer protection and community-level requirements.
The final scope, eligibility thresholds, technical approval criteria, and implementation timetable remain open questions until Ontario publishes its completed framework and any accompanying regulation.
Key Points
- 1Ontario's proposal links new large data centre grid access to government approval, making interconnection policy a material AI infrastructure constraint.
- 2A proposed rate class above 1 MW aims to protect existing ratepayers, potentially changing cost models for GPU-heavy facilities.
- 3The proposal highlights power costs, connection approvals, and community benefits as factors in AI infrastructure development.
Scoring Rationale
Ontario is proposing consequential rules for large AI-oriented data centre connections, electricity cost allocation, and project approvals in a major North American market. The consultation does not create final obligations yet, but it is notable for practitioners and operators whose AI compute roadmaps depend on power availability and grid interconnection.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems


