Operationalizing Responsible, Human-Centered AI
This video is featured in the AI and UX playlist.
Summary
AI-enabled systems that are responsible and human-centered, will be powerful partners to humans. Making systems that people are willing to be responsible for, and that are trustworthy to those using them enables that partnership. Carol will share guidance for operationalizing the work of making responsible human-centered AI systems based on UX research. She will share methods for UX teams to support bias identification, prevent harm, and support human-machine teaming through design of appropriate evidence of system capabilities and integrity through interaction design. Once these dynamic systems are out in the world, critical oversight activities are needed for AI systems to continue to be effective. This session will introduce each of these complex topics and provide references for further exploration of these exciting issues.
Key Insights
-
•
Responsible AI systems require humans to retain ultimate responsibility and control.
-
•
Bias in AI data is inevitable, but awareness and mitigation of harmful bias is crucial.
-
•
Human-machine teaming must be designed with clear responsibilities and transparency.
-
•
AI systems are dynamic and constantly evolving, making continuous oversight essential.
-
•
Speculative exercises like 'What Could Go Wrong' support anticipating harms proactively.
-
•
Calibrated trust in AI means users neither overtrust nor undertrust the system.
-
•
Ethical frameworks such as the Three Q Do No Harm help plan for impact on vulnerable groups.
-
•
Diverse teams improve innovation by being more aware of biases and ethical variation.
-
•
UX practitioners should understand AI concepts to effectively contribute without needing deep technical skills.
-
•
Designing safe AI includes making unsafe actions difficult and safe states easy to maintain.
Notable Quotes
"Responsible systems are systems that keep humans in control."
"Data is a function of our history; it reflects priorities, preferences, and prejudices."
"AI will ensure appropriate human judgment, not replace it."
"We want people to gain calibrated levels of trust, not overtrust or undertrust."
"If the system is not confident, it should transparently communicate that and hand off to humans."
"Ethical design is not superficial; if we don't ask the tough questions, who will?"
"We need to be uncomfortable and get used to asking hard questions about AI."
"Humans are still better at many activities and those strengths should be prioritized."
"Adopting technical ethics gives teams permission to question implications beyond opinions."
"These systems aren’t stable like old software; they change as data and models evolve."
Or choose a question:
More Videos
"Empathy in jazz means the band is in it together—when someone plays a wrong chord, the rest adapt and turn it into an opportunity."
Jim KalbachJazz Improvisation as a Model for Team Collaboration
November 6, 2017
"Writing a book is awful in terms of time, sweat, and time away from your loved ones, but it’s an amazing gift to dig deep into something you care about."
Louis RosenfeldCoffee with Lou: Should You Write a (UX) Book?
March 7, 2024
"If you’re sitting there thinking I need to build out this project and nobody else is saying the same thing, it means you have to do it, period."
Catt Small Micah Bennett Brian Carr Jessica HarlleeWhat's Next for ICs: Exploring Staff and Principal Designer Roles
February 22, 2024
"You might find that spending the most time getting people interested in collaboration is where the work really happens."
Marieke McCloskeyUser Science: Product Analytics & User Research
March 11, 2021
"DeepFakes can look too realistic and might introduce new privacy risks if donor faces come from real people."
Llewyn Paine[Demo] Deploying AI doppelgangers to de-identify user research recordings
June 5, 2024
"Random assignment to treatment or control is essential—if it’s not random, it’s not a true experiment."
Joshua NobleCasual Inference
October 6, 2023
"Choosing the right output depends on the desired outcome—from quick iterations to gaining senior leadership buy-in."
Sara LogelYour Colleagues are Your Users Too
March 29, 2023
"We are obligated to treat each other with kindness and respect; this is a safe and inclusive space."
Bria Alexander Louis RosenfeldWelcome
January 8, 2024
"Timely interactions that log users out without saving progress cause abandonment, especially for people with disabilities."
Sam ProulxOnline Shopping: Designing an Accessible Experience
June 7, 2023
