Skip to content
This is NOT an official site of the Government of Canada. Click here for the official AI registry.

Helpful Places

Government of Canada – DTPR AI Register

AI-Assisted Product Label Compliance Inspection

Enforcement

What it collects

About a measurement
Anonymized data
  • Images of product labels uploaded by Health Canada inspectors via the web application, containing ingredient lists, product names, and other labelling information. The register confirms no personal information is involved.
Operational data
Anonymized data
  • The Cosmetic Ingredient Hotlist, a publicly published regulatory reference list maintained by Health Canada identifying restricted and prohibited cosmetic ingredients. This serves as the reference dataset for compliance validation.
Run by
Health Canada (HC)
Where
No fixed location
Kept
Not stated by the Helpful Places.
Shared with
Accountable organization
Your copy
You cannot see the data it holds about you. What you can do

What it is for

Enforcement

This system helps Health Canada inspectors verify whether product labels comply with federal regulations. Inspectors upload images of product labels, and the tool uses optical character recognition to extract ingredient lists and checks them against Health Canada's Cosmetic Ingredient Hotlist of restricted and prohibited substances. The system then generates a compliance report. Only government employees use this tool; it does not directly affect individual consumers.

What it collects and what happens to it

Data taken in

About a measurement
Anonymized data
  • Images of product labels uploaded by Health Canada inspectors via the web application, containing ingredient lists, product names, and other labelling information. The register confirms no personal information is involved.
Operational data
Anonymized data
  • The Cosmetic Ingredient Hotlist, a publicly published regulatory reference list maintained by Health Canada identifying restricted and prohibited cosmetic ingredients. This serves as the reference dataset for compliance validation.

Processing

Classification & Prediction
  • OCR-extracted ingredients are classified against a rule-based restricted/prohibited list using string-matching or similar lookup techniques. This is a deterministic validation step rather than a probabilistic machine learning model, but functions as a classification within the pipeline.

What it does

Sensing (Perceptive AI)
Human decides
  • The system uses optical character recognition (OCR) to read product label images uploaded by inspectors, extracting the list of ingredients as structured text for downstream validation.
Deciding (Analytical AI)
Human decides
  • The system classifies extracted ingredients as compliant or non-compliant by validating them against the Cosmetic Ingredient Hotlist (restricted and prohibited substances lists). A human inspector reviews the generated compliance report before any enforcement action is taken.

Outputs

A recommendation or prediction
Anonymized data
  • A generated compliance report indicating whether the product label's ingredients are compliant, restricted, or prohibited under Health Canada regulations. The report is advisory — a human inspector uses it to inform enforcement decisions.

Run by

Health Canada (HC)
  • Health Canada is the federal department responsible for helping Canadians maintain and improve their health. It is accountable for the deployment and oversight of this AI-assisted label compliance inspection tool.

GC AI Register — Project Cyclops (2526-HC-SC-002)

Built by

Cistel Technology
  • Cistel is the vendor contracted to build and supply the Project Cyclops AI system for Health Canada.

GC AI Register — Project Cyclops (2526-HC-SC-002)

Kept for

Not stated by the Helpful Places.

Shared with

Not available to me
  • The system is used exclusively by Government of Canada employees (Health Canada inspectors). Members of the public and affected product manufacturers do not have access to the tool or its outputs.
Available to the accountable organization
  • Compliance reports and processed data are available to Health Canada inspectors and the department as part of the inspection workflow.

Stored

Not stated by the Helpful Places.

How to read the colours

Can it identify you?

Anonymized data
Data about people with the link to who is broken. Stripped of identifiers, blurred, aggregated, or noised so this system can’t reasonably tie a record back to an individual.
Pseudonymous data
Each person’s data is tied to a token (hash, ID, template) that lets this system recognise the same person across events, but the token itself doesn’t reveal a name. Reidentification is possible with extra information.
Identifiable data
The data either contains a direct identifier (name, address, account name, recognisable face or voice, plate number) or carries a token this system uses to look up legal identity during processing.

Who completes the loop?

Human decides
This mode suggests; a person decides what to do next. The AI is always advisory — a human is in the loop on every decision. Example: a triage tool ranks cases for a clinician who chooses which to see first.
Human executes
This mode decides; a person carries out the result. Example: an optimizer plans the day’s trash-collection routes, and drivers run them.
Autonomous
This mode decides and acts on its own. No person reviews each decision or carries out the resulting action.

Definitions from the DTPR standard. Amber is about your data, violet about who decides. The fuller the shape and the deeper the colour, the more identifying the data or the less a person is involved.

What you can do

Ask about this system

Questions go to the Helpful Places, not the vendor.

Your rights

  • Right to Be Informed of AI UseThe official register records that AI use is not currently disclosed to users (AI use disclosed to users: N). Product manufacturers whose labels are inspected using this tool are not informed that an AI system was involved in the compliance assessment. Members of the public may consult the Government of Canada's AI Register at open.canada.ca to learn about this system.

Risks and safeguards

  • Reputational harmOCR errors or false positives in ingredient matching could incorrectly flag a compliant product as non-compliant, potentially harming a manufacturer's reputation or triggering unwarranted enforcement actions.Safeguard: the system is in development and the compliance report is advisory, with a human inspector reviewing outputs before any enforcement action. The system is used by trained GC employees rather than being exposed to the public directly.