Implementing Ethics: Developing Trustworthy AI
summary
Claims a reusable UX framework with four concrete checks: accountable to humans, cognizant of speculative risks and benefits, respectful and secure, and honest and usable. Demonstrates a working prototype called Right Staff (AI shift scheduler for store managers), shows how to let managers override schedules, log and appeal decisions, and plan mitigation for reinforcement of bias. Uses a downloadable ethics checklist, Montreal Declaration as the chosen charter, and UX exercises like Black Mirror episodes to find abuse modes.
AI-written summary · about
chapters (24)
- 0:00Intro
- 0:25Early, purposeful work
- 0:40Start with a diverse team
- 1:35Curiosity Experts
- 2:25High value in diverse teams
- 3:24Inclusive Workplace Requirements
- 4:18Adopt Technology Ethics
- 5:43Coalesce on Shared Set of Technology El
- 7:01Conversations for Understanding
- 8:01Prompt conversations with a checklist
- 9:21UX Framework for Designing Trustworthy
- 9:36RightStaff Scenario
- 10:17Accountable to Humans
- 10:40Ensure humans can unplug the machine
- 11:05Significant decisions
- 11:36Responsibilities explicitly defined
- 12:55Cognizant of Speculative Risks and Benell
- 13:23Speculative: Conduct UX research and activate curiosity
- 14:13Potential Abusability of RightStaff
- 14:57Speculative: Create communication & mitigation plans
- 15:41Respectful and Secure
- 17:11Fair: Remove unwanted bias in data
- 18:06UX Framework: Being intentional, keeping people safe
- 18:58Reward team members for finding ethics bugs
related talks
32mSo you want to develop ethical AI. What does that mean in practice?
34mResponsible AI: The Key to Ethical, Safe and Inclusive Software Development
47mHow can we make the most of Ethics applied to AI?
03mDeveloping AI responsibly
40mSocially responsible AI: Practical tools for robust design
27mWriting the Playbook for Fair & Ethical Artificial Intelligence & Machine Learning
36mKeynote: Ethics in AI: Building Fair and Just Intelligence Systems
45mEthical use of AI: seeking, understanding, and countering the risks