Rosenverse

This video is only accessible to Gold members. Log in or register for a free Gold Trial Account to watch.

Log in Register

Most conference talks are accessible to Gold members, while community videos are generally available to all logged-in members.

[Demo] Deploying AI doppelgangers to de-identify user research recordings
Gold
Wednesday, June 5, 2024 • Designing with AI 2024
Share the love for this talk
[Demo] Deploying AI doppelgangers to de-identify user research recordings
Speakers: Llewyn Paine
Link:

Summary

Under biometric privacy laws like BIPA and CCPA, user research recordings containing users’ faces or voices can put your company at risk for lawsuits and fines. Legal departments are increasingly requiring more stringent redaction, and in some cases banning recording outright. This comes at a high cost for UX teams who are already being asked to do more with less, as losing access to recordings can increase duplicative research effort and reduce the accuracy of results. AI offers new solutions for UX teams who want to keep research recordings longer without violating biometric privacy laws. In this demo, we’ll show how we used off-the-shelf tools to intelligently redact users’ voices, faces, and bodies in research videos. By removing biometric identifiers, you can compliantly archive research recordings indefinitely, enabling your team to mine them for insights for years to come.

Key Insights

  • Biometric privacy laws increasingly restrict storing videos with real user faces and voices, creating risks for traditional UX research recordings.

  • Replacing users in research videos with AI-generated avatars preserves rich contextual information without violating privacy rules.

  • Tools like 11 Labs can synthesize user voices, and Wonder Studio can replace entire user bodies in videos using AI-driven 3D avatars.

  • Replacing the whole body rather than just the face helps avoid future biometric identification risks from unique markers like tattoos or moles.

  • Current AI tools for avatar creation have technical limits such as processing time restrictions and unclear licensing for enterprise legal compliance.

  • Blurring or redacting faces is inadequate for privacy as it loses context and doesn't protect voice biometric data.

  • Synthetic avatars raise ethical questions about authenticity and the accuracy of observed user behavior in research videos.

  • Removing visual user identity elements risks erasing diverse perspectives or accountability from the research process.

  • There is an ongoing tension between legal privacy obligations and maintaining the value of user research artifacts.

  • Widespread adoption of AI avatar tools in UX research will require collaboration with legal teams to balance innovation, privacy, and ethics.

Notable Quotes

"If you work in UX, user recordings are arguably your most valuable asset."

"Recording user research is becoming riskier due to growing biometric privacy laws in the US and Europe."

"Replacing entire bodies with avatars sidesteps concerns about unique identifiers like tattoos or moles."

"DeepFakes can look too realistic, so for legal reasons we want it clear at a glance this is not a real person."

"Blurring the entire frame loses a lot of information and still doesn’t do anything about the voice."

"Wonder Studio uses AI to segment an actor, understand their movements, and map those to a 3D model automatically."

"These tools give UX teams options to address legal concerns without sacrificing their most valuable design asset."

"How do we know what’s real if you can’t see a human in a video? Could the whole conversation be fabricated?"

"Removing visual aspects of identity can make it easier to exclude certain groups from our design process."

"Ultimately, what you can do with user recordings is a negotiation with your legal department’s risk tolerance."

Ask the Rosenbot
Mark Interrante
Collaboration Flows in Product Development
2017 • Enterprise Experience 2017
Gold
Kayla Farrell
What It's Like To Be a User Researcher at Compass
2021 • Advancing Research 2021
Gold
Matt Stone
Scaling Empathy, A Case Study in Change Management
2021 • Design at Scale 2021
Gold
John Maeda
About Design Organizations (Videoconference)
2019 • DesignOps Community
Nicole Aleong
What UX research can learn from other research practices [Advancing Research Community Workshop Series] (Videoconference)
2023 • Advancing Research Community
Dave Malouf
Closing Keynote: Amplify. Not Optimize.
2019 • DesignOps Summit 2019
Gold
Cennydd Bowles
Day 1 Panel
2024 • Designing with AI 2024
Gold
Kit Unger
Theme 3 Intro
2022 • Design at Scale 2022
Gold
Bria Alexander
Theme 1 Intro
2024 • DesignOps 2024
Gold
Alla Weinberg
Design Teams Need Psychological Safety: Here’s How to Create It
2022 • DesignOps Summit 2022
Gold
Dr. Jamika D. Burge
How UX researchers can partner with (and not be replaced by) AI [Advancing Research Community Workshop Series](Videoconference)
2023 • Advancing Research Community
Marina Martin
Lives on the Line: The Stakes of UX at the Scale of Government
2018 • Enterprise Experience 2018
Gold
Scaling Design through Relationship Maps
2017 • DesignOps Summit 2017
Gold
Jen Cardello
Learning Velocity—The Insights Speedometer (Videoconference)
2021 • Advancing Research Community
Laureen Kattan
Centering Patients and Clinicians in a Complex Government Ecosystem
2023 • Design in Product 2023
Gold
Louis Rosenfeld
Welcome / Housekeeping
2023 • Enterprise UX 2023
Gold

More Videos

Jack Moffett

"There’s often a disconnect because the people doing the research and conceptualization aren’t the same as those doing the development work."

Jack Moffett

SAFe or Sorry? (Videoconference)

May 29, 2019

Tricia Wang

"Being comfortable with discomfort is a great way to stick with it, especially as someone who might be called out or told they’re wrong."

Tricia Wang

SCALE: Discussion

June 15, 2018

Fredrik Matheson

"Great user experience is not a goal. You need specific, measurable goals like reducing training time by four."

Fredrik Matheson

First-time users, longtime strategies: Why Parkinson’s Law is making you less effective at work – and how to design a fix.

June 8, 2016

Laine Riley Prokay

"Competencies allow managers to celebrate individualism while ensuring consistency."

Laine Riley Prokay

How DesignOps can Drive Inclusive Career Ladders for All

September 30, 2021

Shipra Kayan

"Our NPS plummeted after a rebrand and that freaked out executives—it was our fire moment to start VOC."

Shipra Kayan

How we Built a VoC (Voice of the Customer) Practice at Upwork from the Ground Up

September 30, 2021

Kristen Guth, Ph.D.

"The definition of intuition is knowing without knowing how you know, I just know it."

Kristen Guth, Ph.D.

Out of the FOG: A Non-traditional Research Approach to Alignment

March 28, 2023

Joi Freeman

"Often diverse candidates say they have to keep their heads down just to fit in and move up."

Joi Freeman

A New Vantage Point: Building a Pipeline for Multifaceted Research(ers)

March 30, 2020

Victor Udoewa

"Listening in real-time to user feedback lets you pivot before it’s too late."

Victor Udoewa

Theme One Intro

March 27, 2023

Peter Merholz

"Strong design leadership manages down, across, and up, enabling teams to do their best work."

Peter Merholz

Customer-Centered Design Organizations

June 8, 2017