AI Triage for Employment Insurance Benefit Recalculations

The Employment Insurance Recalculation Triage Machine Learning System is a tool developed by Employment and Social Development Canada (ESDC) to help manage a growing backlog of Employment Insurance (EI) benefit recalculation cases. Each year, approximately 300,000 recalculation work items are generated when new information or changes in circumstances require a review of an existing EI claim. This system uses a machine learning model to predict the likely outcome of each recalculation — whether it will result in an increase, decrease, or no change to a claimant's benefits — so that human officers can focus their attention on cases most likely to affect claimants.

If your EI claim is in the recalculation queue, this system may classify your case as low-priority (no anticipated change) or high-priority (likely change in benefits). Cases classified as having no change may be closed without a full manual review, while cases flagged as having a likely benefit change will be prioritized for officer attention. The system achieves approximately 90% accuracy in identifying no-change claims. Human officers retain the ability to override any system decision, and if you contact Service Canada about your file, an officer will proceed with a full recalculation regardless of the system's classification.

This system operates under the Treasury Board of Canada Secretariat's Directive on Automated Decision-Making and has undergone an Algorithmic Impact Assessment. A Privacy Impact Assessment has been completed, and personal information is used only to the extent necessary to deliver the EI program. The system uses a primary key de-identification method to protect personal data and operates within a closed system with no internet connectivity. If you believe a decision affecting your EI benefits is incorrect, a recourse process is available to challenge it.