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

Back to your search

AI-Assisted Drug and Health Product Information Extraction

Healthcare · Research & Development

What it collects

Operational data
Anonymized data
  • Drug Product Database (DPD) records, Notice of Compliance (NOC) database entries, and Product Monograph (PM) PDF files. These are regulatory and administrative datasets with no personal information.

Government of Canada AI Register — Dark Matter

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

HealthcareResearch & Development

Dark Matter is a database system used by Health Canada to store results from machine learning models that extract and link information from drug product monographs and related health databases. It allows AI systems to identify important documents and find relationships between different types of text. The system is used exclusively by Government of Canada employees and does not involve personal information. Users can also review and correct errors made by the machine learning models.

What it collects and what happens to it

Data taken in

Operational data
Anonymized data
  • Drug Product Database (DPD) records, Notice of Compliance (NOC) database entries, and Product Monograph (PM) PDF files. These are regulatory and administrative datasets with no personal information.

Government of Canada AI Register — Dark Matter

Processing

Classification & Prediction
  • Machine learning models classify and extract structured information from unstructured drug product monograph text, including relationships between text types and inference of document importance. Results are stored in the Dark Matter database for downstream use.

Government of Canada AI Register — Dark Matter

What it does

Understanding (Semantic AI)
Human decides
  • The system uses machine learning to understand and link text across drug monographs and health product databases, finding relationships between different types of text and inferring document importance. Human users review and can correct the model's outputs.

Government of Canada AI Register — Dark Matter

Deciding (Analytical AI)
Human decides
  • Machine learning models predict and classify information extracted from product monographs, scoring or ranking document elements for relevance. Outputs are stored in the database and are subject to human review and correction.

Government of Canada AI Register — Dark Matter

Outputs

Operational data
Anonymized data
  • Structured extraction results from drug monographs — classified text segments, inferred relationships between drug product data, importance scores, and links to Drug Product Database and natural health product records. No personal information is produced.

Government of Canada AI Register — Dark Matter

Run by

Health Canada (HC)
  • Health Canada is the federal department responsible for this AI system. The system is developed and operated by the Government of Canada for use by GC employees.

Government of Canada AI Register — Dark Matter

Built by

Government of Canada
  • The system was developed internally by the Government of Canada, with no external vendor identified in the register entry.

Government of Canada AI Register — Dark Matter

Kept for

Not stated by the Helpful Places.

Shared with

Available to the accountable organization
  • The system is accessible to GC (Government of Canada) employees, specifically within Health Canada. Data export capabilities are available to these internal users.

Government of Canada AI Register — Dark Matter

Not available to me
  • The system is restricted to Government of Canada employees. Members of the public do not have access to the database or its outputs. Additionally, AI use is not disclosed to end users.

Government of Canada AI Register — Dark Matter

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 register states that AI use is not disclosed to users of this system. This means individuals whose drug product information is processed by this system are not informed that AI is being used. This is a transparency gap noted in the official register entry.
  • Right to Correct Your DataThe system includes a built-in mechanism for GC employees to review and correct errors made by the machine learning models. This correction capability is available to internal users, though no public-facing correction process is described.

Risks and safeguards

  • Reputational harmMachine learning extraction errors could result in incorrect associations between drug products, potentially affecting regulatory decisions.Safeguard: the system allows users to identify and fix possible errors in detection done by ML models before results are used downstream.