Week 2: Core Ethics

<aside>

🧭 Dashboard · Week 1 **** · Week 2 · **Week 3 ·** Week 4 · Resource Library

</aside>

Bias, Privacy & Ownership - Week 2 — May 29 · 3:00–4:30 PM PT

</aside>

Selected Week 2 thumbnail.jpeg

https://www.youtube.com/watch?v=rzeE0uBH2gI&feature=youtu.be

Theme

The Three Pillars

Every AI system is a social artifact. It reflects the values of those who built it, the data it was trained on, and the incentives that shaped its design. This week we examine the three ethical fault lines that show up most reliably in practice: bias, privacy, and ownership.

Bias & Prediction: Identify algorithmic bias, understand fairness definitions
Privacy & Consent: Data collection, consent, transparency, and responsible use
Copyright & Creative Work: Who owns AI inputs and outputs? Are we going to use copyright as the 3rd pillar? Do we have enough content?

Pre-reading

Participants should arrive with a working sense of the three pillars and at least one real example from their own context in mind.

Bias & Fairness

Name the Bias: Kris Krug, 2026 - A practitioner's lens on identifying and naming AI bias in real contexts.

Name the Bias - Kris Krüg | Generative AI Tools & Techniques

Read →

</aside>

Ensuring Trustworthy AI through AI Testing (Fundamentals): IBM - Spend 10 minutes exploring how fairness metrics work in practice. Hands-on entry point into bias measurement and mitigation. AI application testing fundamentals.

Ensuring Trustworthy AI through AI Testing

Explore →

</aside>

Data Ethics Canvas: Open Data Institute - A plain-language worksheet [direct PDF download link] for ethical review of data projects.

The Data Ethics Canvas

Explore →

</aside>

DWP algorithm wrongly flags 200,000 people for fraud: The Guardian, 2024 - A real-world example of what goes wrong when algorithmic outputs go unexamined.

DWP algorithm wrongly flags 200,000 people for possible fraud and error

Read →

</aside>

Human attention and AI: MIT News, 2026 - Interactive data visualization tools that prioritize human agency. Explores how design shapes attention and decision-making.

A human-centered approach to data visualization

Read →

</aside>

Privacy & Consent

Canada's Algorithmic Impact Assessment: Government of Canada - A 65-question self-assessment for AI systems. Skim the structure to understand the risk-tier approach.

Algorithmic Impact Assessment tool

Explore →

</aside>

PIPEDA in plain language: Office of the Privacy Commissioner of Canada - A 5-minute read on Canada's federal privacy baseline.

PIPEDA fair information principles

Read →

</aside>

Privacy Laws in BC: PIPA - BC Laws - The Personal Information Protection Act — BC's provincial privacy standard.

Personal Information Protection Act

Read →

</aside>