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 Writing and Productivity Assistant for Correctional Staff

Employment & Work

What it collects that can identify you

Sensitive personal information
Identifiable data
  • The system involves personal information, as declared in the register. This likely includes content that employees work with in Microsoft 365 applications, which may contain personally identifiable information related to government operations or individuals under correctional supervision.

Also collects operational data, which is anonymized data.

Run by
Correctional Service of Canada (CSC)
Where
No fixed location
Kept
No data retained
Shared with
Accountable organization

What it is for

Employment & Work

The Correctional Service of Canada is piloting Microsoft 365 Copilot, an AI-powered assistant that helps government employees draft documents, summarize information, and complete everyday office tasks. The pilot is limited to 50 licences and is still in development, with security and IT guidelines being established before wider rollout. The system involves personal information and users are informed that AI is in use.

What it collects and what happens to it

Data taken in

Sensitive personal information
Identifiable data
  • The system involves personal information, as declared in the register. This likely includes content that employees work with in Microsoft 365 applications, which may contain personally identifiable information related to government operations or individuals under correctional supervision.
Operational data
Anonymized data
  • Employees provide documents, emails, and other workplace content from within the Microsoft 365 environment as prompts or context for the AI assistant. The tool does not retain context across multiple sessions.

Processing

Language Models
  • Microsoft 365 Copilot is built on large language model technology integrated within the Microsoft 365 application suite. It uses language models to generate, summarize, and transform text in response to user prompts.

What it does

Creating (Generative AI)
Human decides
  • Microsoft 365 Copilot generates text, summaries, and document drafts in response to prompts from government employees. A human employee reviews and decides whether to use the generated output.
Understanding (Semantic AI)
Human decides
  • The system reads and comprehends documents and data within the Microsoft 365 environment to retrieve relevant context, summarize content, and surface related information for employee use.

Outputs

Generated content
Anonymized data
  • The system produces AI-generated text, document drafts, summaries, and other written content for review and use by government employees in their day-to-day work.

Run by

Correctional Service of Canada (CSC)
  • A federal government department that is piloting Microsoft 365 Copilot for use by its employees, under a 50-licence limited rollout. It is establishing cyber and IT security guidelines to govern responsible use of the system.

Government of Canada AI Register — 2526-CSC-SCC-001

Built by

Microsoft
  • Microsoft is the vendor that builds and supplies Microsoft 365 Copilot, the AI system being piloted by the Correctional Service of Canada.

Government of Canada AI Register — 2526-CSC-SCC-001

Kept for

No data retained
  • The tool does not retain context across multiple sessions, indicating that session-level data is not persisted by the AI component after each interaction ends.
  • Duration: Not retained across sessions

Shared with

Available to the accountable organization
  • Outputs generated by the system are available to the Correctional Service of Canada employees who use the tool. The register does not specify broader data-sharing arrangements.

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 UseAI use is disclosed to users. Government employees using Microsoft 365 Copilot are informed that they are interacting with an AI system. Contact the Correctional Service of Canada for further information on how disclosure is implemented.
  • Right to Algorithmic TransparencyThis system is listed in the Government of Canada's AI Register, providing baseline transparency about its existence, purpose, and data practices. Further details about the underlying model logic are governed by Microsoft's published documentation for Microsoft 365 Copilot.

Risks and safeguards

  • Civil liberties harmBecause the system operates in a correctional context and involves personal information, there is a risk that AI-generated content could influence decisions affecting individuals under correctional supervision, potentially implicating due process rights.Safeguard: The pilot is limited to 50 licences, is still in development, and CSC is establishing cyber and IT security guidelines before broader deployment. Users are disclosed the AI nature of the tool.
  • Psychological harmAI-generated content presented as authoritative could inadvertently cause distress or anxiety if relied upon for decisions affecting individuals in the correctional system.Safeguard: The tool is used by GC employees who retain decision-making authority; outputs are advisory and require human review before any consequential use.