City aims to catch permit problems sooner with new AI screening tool
Greater Sudbury council has approved a $800,000 pilot project that will introduce an artificial intelligence system to screen building permit applications for missing information and potential conflicts with city zoning rules and parts of the Ontario Building Code. Council endorsed the proposal at its Tuesday afternoon meeting, with staff saying the technology is intended as an initial screening tool rather than a substitute for human reviewers.
The pilot will operate through the city's online submission system, Pronto. When permit plans are uploaded, the AI software will perform an initial review to identify incomplete applications and flag potential zoning or code-related issues so applicants can address them sooner. City officials say that could reduce the time applicants currently wait for an initial human review.
Kris Longston, general manager of planning and growth for the city, told council the system will first be trained on Greater Sudbury’s zoning rules and the municipality’s building-permit process. He compared the technology to "hiring a new employee" that needs time to learn the system before tackling more complex tasks. Longston also said he expects the tool eventually could stop incomplete applications from being submitted through Pronto.
- Purpose: Automated screening for missing information and potential zoning/building-code issues.
- Scope: Initial review only; complex assessments and final decisions remain with city staff.
- Platform: Integrated with the Pronto permit-submission portal.
The city’s description of the project emphasises speed and early error detection. Currently, applicants may wait several days for an initial staff review to learn that their submission is incomplete or requires changes. With the AI screening in place, applicants could receive near-immediate feedback, shortening the cycle for corrections and resubmission.
Officials say the system will be trained on municipal rules and will evolve over time. The plan envisions a staged rollout where the technology begins by handling straightforward checks and gradually takes on more of the routine screening work as its accuracy improves. Throughout, municipal staff will continue to perform in-depth reviews and issue final rulings on permits.
"The system would first be trained on the city's zoning rules and building permit process. Longston compared it to hiring a new employee who needs to learn how the system works before taking on more complicated tasks."
City staff framed the pilot as a way to improve service delivery without replacing existing expertise. The intent is to reduce wait times for applicants — homeowners, contractors and developers — while keeping technical, discretionary and safety-related judgments under professional human control.
| Item | Role |
|---|---|
| AI software | Initial screening for completeness and code/zoning flags |
| City staff | Complex reviews, discretionary decisions, final permit approvals |
| Pronto portal | Submission point for applications; integration with AI screening |
Introducing automation to municipal permit processes raises questions about accuracy, fairness and accountability in public administration. The city’s publicly stated approach places those responsibilities with staff while using technology to surface routine issues earlier. That balance will be watched closely by local builders and homeowners if the pilot proceeds as planned.
For Sudbury, a city with a mix of dense urban neighbourhoods and sprawling rural areas, reducing permit turnaround times could have practical effects on small contractors working across municipal wards and on residents undertaking renovations or new builds. Faster initial feedback on permits can limit costly delays and reduce the administrative burden of multiple resubmissions.
Council’s approval of the pilot marks the first formal step toward integrating AI into Greater Sudbury’s building-permit workflow. The coming weeks should reveal the project timeline, vendor selection and metrics the city will use to judge whether the technology meets goals for speed, accuracy and public service.
