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 Policy Marker Assessment for International Aid Projects

Eligibility & Public Benefits · Planning & Decision-making

What it collects

About behaviour
Anonymized data
  • Project summaries entered by GAC staff into the web-based application. These summaries describe international assistance project activities and objectives, and are generated during GAC's existing grant and contribution management processes. The register confirms no personal information is involved.
Operational data
Anonymized data
  • Policy assessment frameworks and domain knowledge encoded within the system, derived from manual assessments used as a gold standard during development. These represent institutional knowledge about Canada's international assistance policy objectives.
Run by
Global Affairs Canada (GAC)
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

Eligibility & Public BenefitsPlanning & Decision-making

This tool uses generative AI to automatically assign policy markers to international assistance project summaries submitted by Global Affairs Canada staff. It encodes the assessment frameworks and domain knowledge of experienced evaluators to help staff quickly determine how projects align with policy objectives. The system produces an explanation alongside each marker assignment so that experts can review and verify the results. It does not process personal information.

What it collects and what happens to it

Data taken in

About behaviour
Anonymized data
  • Project summaries entered by GAC staff into the web-based application. These summaries describe international assistance project activities and objectives, and are generated during GAC's existing grant and contribution management processes. The register confirms no personal information is involved.
Operational data
Anonymized data
  • Policy assessment frameworks and domain knowledge encoded within the system, derived from manual assessments used as a gold standard during development. These represent institutional knowledge about Canada's international assistance policy objectives.

Processing

Language Models
  • Generative AI methods are used to encode assessment frameworks and evaluator domain knowledge, enabling the system to read project summaries in natural language and assign policy markers with explanations comparable in accuracy to expert human assessors.
Classification & Prediction
  • Classifies project summaries against multiple policy marker categories, selecting the most relevant markers. Accuracy is described as comparable to expert assessors, validated against manual assessments used as a gold standard.

What it does

Creating (Generative AI)
Human decides
  • Uses generative AI to encode assessment frameworks and the domain knowledge of experienced evaluators, producing policy marker assignments and accompanying explanations from project summaries. Expert staff review and validate the outputs.
Deciding (Analytical AI)
Human decides
  • Classifies and ranks project descriptions against multiple policy markers, selecting the most relevant markers for each project. Outputs are advisory — human experts retain final decision authority.

Outputs

A recommendation or prediction
Anonymized data
  • Policy marker assignments and accompanying explanations for each project summary. These are advisory outputs — staff can review the reasoning and override or validate the suggested markers. The register confirms no personal information is produced.
Generated content
Anonymized data
  • Natural-language explanations generated by the AI to accompany each policy marker assignment, enabling expert oversight and review of the system's reasoning. The content explains why specific markers were selected for a given project description.

Run by

Global Affairs Canada (GAC)
  • Global Affairs Canada is the federal department responsible for managing most of Canada's international assistance funding. It developed and deploys this AI tool to automate policy marker assessments for grants and contributions projects.

GC AI Register — 2526-GAC-AMC-006

Built by

Government of Canada
  • The system was developed internally by the Government of Canada, with Global Affairs Canada as the primary developer and owner of the solution.

GC AI Register — 2526-GAC-AMC-006

Kept for

Not stated by the Helpful Places.

Shared with

Available to the accountable organization
  • Outputs — policy marker assignments and explanations — are accessible to Global Affairs Canada staff (GC employees) who use the web-based tool. The register identifies primary users as GC employees only.
Not available to me
  • The tool is an internal government application for GC employees only. Members of the public and affected project applicants do not have access to the tool's outputs or the policy marker assignments it generates.

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 Algorithmic TransparencyGlobal Affairs Canada has disclosed AI use to users of this tool. The system produces an accompanying explanation with each policy marker assignment to allow expert oversight and to make the AI's reasoning legible to reviewing staff.
  • Right to a Human ReviewThe tool is currently in validation and is designed for expert oversight. Expert staff review and can override or validate policy marker assignments produced by the AI. The system is explicitly advisory, with human evaluators retaining authority over final assessments.

Risks and safeguards

  • Reputational harmInaccurate policy marker assignments could mischaracterize projects, affecting funding decisions or the reputations of project proponents.Safeguard: The system's accuracy was validated against manual assessments as a gold standard; the tool produces explanations to enable expert review; the system is currently in validation and human experts retain oversight authority over all outputs.
  • Loss of autonomyAutomated policy marker assignments could create anchoring bias, nudging expert reviewers toward accepting AI outputs without independent judgment.Safeguard: The system is designed as explicitly advisory; explanations are provided with each assignment to support independent expert evaluation; the tool is in validation phase with human oversight built into the workflow.