Broad Strokes: Connecting Design, Research, and AI to the World Around Us
Summary
AI-based modeling plays an ever-increasing role in our daily lives. Our personal data and preferences are being used to build models that serve up anything from movie recommendations to medical services with a precision designed to make life decisions easier. But, what happens when those models are not precise enough or are unusable because of incomplete data, incorrect interpretations, or questionable ethics? What's more, how do we reconcile the dynamic and iterative nature of AI models (which are developed by a few) with the expansive nature of the human experience (which is defined by many) so that we optimize and scale UX for AI applications equitably? This talk will push our understanding of the sociotechnical systems that surround AI technologies, including those developing the models (i.e., model creators) and those for whom the models are created (i.e., model customers). What considerations should we keep top of mind? How might the research practice evolve to help us critique and deliver inclusive and responsible AI systems?
Key Insights
-
•
AI systems in healthcare often embed racial biases that exacerbate disparities, particularly impacting Black and Hispanic populations.
-
•
Algorithms based on past healthcare spending can unfairly dictate future levels of care, disadvantaging poorer patients.
-
•
AI hiring tools can perpetuate bias by training on historically discriminatory data, reducing trust among candidates.
-
•
Dark patterns in UI design, when combined with AI, may manipulate users into unintended decisions, especially vulnerable groups.
-
•
Deepfakes, created via AI neural networks, pose serious risks of misinformation, reputational harm, and psychological damage.
-
•
Facial recognition software shows significant racial and gender bias, performing worst on darker-skinned females as shown by Joy Buolamwini.
-
•
Biases in AI extend beyond technical factors to include systemic and human biases embedded in organizations and data collection.
-
•
Ethical failures like the Tuskegee Syphilis Study highlight the urgency of integrating ethics and do-no-harm principles into AI research and design.
-
•
Women and minorities remain vastly underrepresented in technical AI roles, limiting the diversity of perspectives in AI development.
-
•
Emerging regulations like GDPR and CPRA hold organizations accountable for ethical data use, reinforcing the responsibility of AI practitioners.
Notable Quotes
"Imagine going to the hospital for an emergency and the doctor doesn’t believe you’re in pain because AI says so."
"Algorithms that adjust care based on race don’t actually help and embed disparities instead."
"Dark patterns are not new in UX, but AI amplifies their impact by deceiving users in new ways."
"Deepfakes can disrupt democratic elections and inflict lasting psychological harm on victims."
"Biases out of sight are biases out of mind — facial recognition misidentifies Black faces leading to false arrests."
"All commercially available facial recognition software perform worse on darker females."
"Doing no harm is a core tenet of user experience research, whether or not you are trained in IRB processes."
"Only 12% of AI researchers globally are women, and 6% of professional software developers are women of color."
"Policy like the California Privacy Rights Act requires us to explain data use and comply with user deletion requests."
"The best transformation happens by asking the right questions, even when the problem feels overwhelming."
Or choose a question:
More Videos
"Decision making here is basically forge trading — people argue and withhold support until they get what they want."
Carl TurnerYou Can Do This: Understand and Solve Organizational Problems to Jumpstart a Dead Project
March 28, 2023
"Don’t send your recruiters to spam LinkedIn. That damages your brand and turns off potential candidates."
Karen PascoeDeveloping Experience Teams and Talent in the Enterprise
June 8, 2016
"Continuous onboarding is like drinking water in sips throughout the day, not a gallon all at once."
Jilanna WilsonDistributed Design Operations Management
October 23, 2019
"Slack channels and weekly meetings encourage cross-contract collaboration, avoiding duplication and sharing learnings across teams."
Shawna Hein Kevin HoffmanCreate a Cohesive Civic Design Practice Across Agency, Vendors, and Contracts
November 17, 2022
"Move fast and break things doesn’t work when you’re designing for cancer patients or disaster victims."
Barb SpantonDoing Work That Matters: A Look Beyond The Idealistic Notion of 'Doing Meaningful Work'
June 10, 2022
"We decided not to chase shiny objects because we wanted to stay true to our mission."
Briana ThomasThe Quiet Force: Uncovering Hidden Leadership in High-Impact Design Teams
September 24, 2024
"Taking ugly notes together without worrying about making things pretty lowers barriers and increases engagement."
Gina MendoliaTherapists, Coaches, and Grandmas: Techniques for Service Design in Complex Systems
December 3, 2024
"Understanding how assistive technology evolves is crucial because it foretells the future of interface design."
Lija Hogan Milan Mijatovic Sam Proulx Louis RosenfeldThree Years Out: Perspectives on the Near-Term Future of User Research
March 15, 2024
"As your design practice matures, your design ops team will reach inflection points, and it’s okay to flip the coin and explore the other side."
John Calhoun Rachel PosmanTwo Sides of the DesignOps Coin: Teams Ops and Product Ops
January 8, 2024
