wiki

Changes

Artificial Intelligence Tip Sheet (view source)

Revision as of 14:18, 24 September 2025

7,641 bytes added ,  Yesterday at 14:18
no edit summary
Line 22: Line 22:  
* [https://www150.statcan.gc.ca/n1/pub/12-586-x/12-586-x2017001-eng.htm Statistics Canada's Quality Assurance Framework]
 
* [https://www150.statcan.gc.ca/n1/pub/12-586-x/12-586-x2017001-eng.htm Statistics Canada's Quality Assurance Framework]
 
* [https://www.priv.gc.ca/en/privacy-topics/privacy-laws-in-canada/the-personal-information-protection-and-electronic-documents-act-pipeda/p_principle/ PIPEDA fair information principles - Office of the Privacy Commissioner of Canada]
 
* [https://www.priv.gc.ca/en/privacy-topics/privacy-laws-in-canada/the-personal-information-protection-and-electronic-documents-act-pipeda/p_principle/ PIPEDA fair information principles - Office of the Privacy Commissioner of Canada]
 +
 +
=== 2. Developing & Updating Policies ===
 +
When creating or changing policies that involve AI, assess its impact on people from the very beginning.
 +
 +
Conduct an Algorithmic Impact Assessment (AIA) to determine the system's risk level. This must include an explicit assessment of the proposed data sources for potential bias.
 +
 +
Incorporate foundational legislation like the Accessible Canada Act, the United Nations Declaration for the Rights of Indigenous People and the Employment Equity Act  in the policy analysis.
 +
 +
Challenge policy assumptions in areas like risk scoring, eligibility determination, and the underlying assumptions of the policy itself that could lead to discriminatory outcomes.
 +
 +
==== Key Resources: ====
 +
 +
* Algorithmic Impact Assessment (AIA) Tool
 +
* Accessible Canada Act
 +
* United Nations Declaration on the Rights of Indigenous Peoples Act
 +
* Employment Equity Act
 +
* Guide to Peer Review of Automated Decision Systems - Canada.ca
 +
 +
=== 3. Designing Programs & Services ===
 +
Embed fairness directly into the architecture of any AI-powered program.
 +
 +
Ensure meaningful human oversight and provide plain-language notices to users explaining how the AI works and how to challenge a decision.
 +
 +
Include systemically marginalized groups in all phases, from initial design to final testing and implementation.
 +
 +
Audit all AI tools for equity, especially internal systems, to ensure they do not perpetuate bias and barriers.
 +
 +
Embed accessibility and bias mitigation throughout design, testing, and implementation.
 +
 +
==== Key Resources: ====
 +
 +
* Directive on Automated Decision-Making
 +
* Policy on Service and Digital
 +
* Accessible and Equitable Artificial Intelligence Systems - Accessibility Standards Canada
 +
* Guide to Peer Review of Automated Decision Systems
 +
 +
=== 4. Procuring Technology & AI Systems ===
 +
Require conformance with accessibility standards in all procurement contracts. This includes both the hardware/software standards and the specific standards for AI.
 +
 +
Require potential vendors to disclose the sources of their training data, their data-cleaning methods, and the steps they took to mitigate bias in their models.
 +
 +
Require conformance with accessibility standards in all procurement contracts, including ICT and AI-specific standards.
 +
 +
Mandate an external, independent peer review for any high-impact AI system before a contract is finalized and before deployment.
 +
 +
==== Key Resources: ====
 +
 +
* CAN/ASC - EN 301 549:2024 Accessibility requirements for ICT products and services (EN 301 549:2021, IDT) - Accessibility Standards Canada
 +
* ASC-6.2 Accessible and Equitable Artificial Intelligence Systems - Accessibility Standards Canada
 +
* Directive on Automated Decision-Making (See section 6.3.3.6 on GBA Plus)
 +
 +
=== 5. Managing & Supervising Teams ===
 +
As a leader, help build your team's capacity to work with AI ethically and inclusively.
 +
 +
Obtain training on bias, equity, and accessible design principles.
 +
 +
Actively engage employees from systemically marginalized groups to gather feedback on AI tools and processes.
 +
 +
=== 6. Working in Human Resources (HR) ===
 +
Exercise caution to prevent AI from creating discriminatory barriers in recruitment, promotion, or talent management.
 +
 +
Do not use AI in hiring or promotion unless:
 +
 +
AI training data was audited and corrected for biases related to gender, race, disability, and other protected grounds.
 +
 +
The system has been independently audited for equity impacts in a Canadian context.
 +
 +
Interfaces are fully bilingual and accessible.
 +
 +
Ensure AI-enabled learning or assessment platforms are barrier-free and have been co-designed with meaningful consultation from systemically discriminated groups.
 +
 +
Conduct an Algorithmic Impact Assessment for any system that automates decisions affecting employees' rights or careers.
 +
 +
==== Key Resources: ====
 +
 +
* Artificial intelligence in the hiring process - Canada.ca
 +
* Enhance fairness and reduce bias in the content of assessment tools - Canada.ca
 +
* Employment Equity Act
 +
* Accessible and Equitable Artificial Intelligence Systems - Accessibility Standards Canada
 +
* Algorithmic Impact Assessment (AIA) Tool
 +
 +
=== 6. Supporting Indigenous Rights & Self-Determination ===
 +
Ensure AI systems respect the rights and data sovereignty of Indigenous Peoples.
 +
 +
Align AI systems with the principles of the the United Nations Declaration on the Rights of Indigenous Peoples Act and the OCAP® Principles (Ownership, Control, Access, and Possession).
 +
 +
Include Indigenous Peoples, leaders, and networks in the design, procurement, and governance of any AI system that may affect them.
 +
 +
Avoid automated systems that could reinforce or create new systemic inequities for Indigenous Peoples.
 +
 +
==== Key Resources: ====
 +
 +
* The First Nations Principles of OCAP®
 +
* United Nations Declaration on the Rights of Indigenous People Act
 +
 +
=== 7. Monitoring, Evaluating & Auditing ===
 +
Continuously assess the real-world impact of AI systems to ensure they remain fair and effective over time.
 +
 +
Assess AI-related impacts using GBA Plus assessments, program evaluations, and privacy and accessibility audits.
 +
 +
Continuously monitor system outputs for unexpected or inequitable results. If an AI system starts flagging a specific demographic at a higher rate, it requires immediate investigation.
 +
 +
Report transparently on AI risks, mitigation efforts, and any updates made to the system.
 +
 +
Establish clear feedback and redress mechanisms so users can challenge an automated decision
 +
 +
==== Key Resources: ====
 +
 +
* Directive on Automated Decision-Making- Canada.ca 6.3,3.6
 +
* Gender-based Analysis Plus (GBA Plus) - Canada.ca.
 +
 +
=== 8. Engaging the Public & Partners ===
 +
Foster public trust through clear communication and meaningful collaboration.
 +
 +
Provide plain language explanations of what AI tools do and how they impact people.
 +
 +
Ensure all outreach is culturally relevant, linguistically accessible, and inclusive of marginalized communities.
 +
 +
Co-design AI systems with systemically marginalized groups and by recognizing that persons with disabilities must be involved in creating policies and services that affect them.
 +
 +
==== Key Resources: ====
 +
 +
* ASC-6.2 Accessible and Equitable Artificial Intelligence Systems - Accessibility Standards Canada
 +
 +
=== 9. Designing Training & Communications ===
 +
Create educational materials that are accessible and inclusive.
 +
 +
Use inclusive and accessible formats like screen-reader-compatible documents, captioned videos, and translated materials.
 +
 +
Co-create content with  systemically marginalized groups to ensure it is relevant and respectful.
 +
 +
Go beyond "performative" training. Invest in meaningful education that helps public servants understand and confront systemic racism, ableism, and colonialism.
 +
 +
==== Key Resources: ====
 +
 +
* ASC-6.2 Accessible and Equitable Artificial Intelligence Systems - Accessibility Standards Canada
 +
 +
=== 10. For All Public Servants: Your Personal Responsibility ===
 +
Regardless of your role, you have a part to play in ensuring the responsible use of AI.
 +
 +
Seek out training on data literacy, AI ethics, and unconscious bias. Understand the basics so you can ask critical questions.
 +
 +
If you are asked to work on an AI project, ask "What are the risks of bias in this data?" and "Who might be negatively impacted by this system?"
 +
 +
Actively consult with systemically marginalized groups and colleagues from different levels, locations and classifications. Their insights are invaluable for spotting potential issues.
 +
 +
==== Key Resources: ====
 +
 +
* AI Strategy for the Public Service
 +
* Guiding principles for the use of AI in government
 +
* Guide on the use of generative artificial intelligence
Retrieved from "https://wiki.gccollab.ca/Special:MobileDiff/138408"