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] Stress-testing GenAI in user research synthesis
Gold
Tuesday, June 4, 2024 • Designing with AI 2024
Share the love for this talk
[Demo] Stress-testing GenAI in user research synthesis
Speakers: Yulya Besplemennova
Link:

Summary

In this high-pressure scenario, the challenge was to conduct 17 user interviews in three days and synthesize a comprehensive report in just one additional day. I’ll explore how we used AI to streamline the research process, from transcription to synthesis, and how tools like ChatGPT contributed to efficient data processing and insight generation. We’ll reflect on the potential and pitfalls of using AI in accelerated user research, from practical aspects to more philosophical considerations on potential changes to the research process. Takeaways Practical insights into integrating AI with traditional research methodologies to expedite the research process An overview of the effectiveness of AI transcription and synthesis tools in real-world research scenarios Critical examination of AI's role in data processing and how it compares with human analysis Strategic considerations for service designers when employing AI to support rapid user research Reflection on the ethical implications and potential impact on the quality of insights and researcher well-being when relying on AI to speed up research processes

Key Insights

  • Generative AI can significantly speed up text generation and help brainstorm clustering parameters in research synthesis.

  • AI struggles to assign specific interviewees to user archetypes without detailed human guidance.

  • The best use of AI involves a semi-automated process combining AI brainstorming with human refinement.

  • Automated visuals generated by DALL·E often fail to accurately represent complex user concepts.

  • AI behaves like an 'alien intern,' lacking common sense and understanding of research processes.

  • Providing AI with explicit details like participant names improves clustering and assignment accuracy.

  • AI can efficiently retrieve specific quotes from transcripts when given open-ended queries.

  • Human experience is crucial to evaluate and guide AI outputs, preventing misleading or irrelevant results.

  • AI reduces cognitive overload and accelerates synthesis during tight deadlines and solo work.

  • Rapid AI-assisted research risks sacrificing depth and nuance of human insights for speed.

Notable Quotes

"Gene AI can really help to speed up some certain synthesis moments, especially those connected to textual content generation and navigation."

"AI behaves kind of like an alien intern because it’s not capable of making sense of some things which are easy for us."

"I had to adapt and follow a semi-automated process where I would brainstorm a lot of things with AI, then define many things by myself."

"The biggest superpower of ChatGPT right now is generating textual descriptions along the given format when it understands it and the relevance."

"Visuals are not working that well with DALL·E in this context, despite multiple instructions to remove cables or irrelevant elements."

"AI can provide multiple inputs that stimulate human thinking even if many are generic and need human filtering."

"You need to guide AI a lot and do handholding, even reminding it simple things like how many people are in the transcript."

"It’s better to have a human, but if you don’t, AI can be a partner in the process to speed things up."

"The value is not so much about the results but about helping manage cognitive overload and speed up the process under tight deadlines."

"AI making the process five times faster is worrying if it means losing the depth of human insights for speed and efficiency."

Ask the Rosenbot
Luke Roberts
Panel Discussion
2024 • Advancing Service Design 2024
Gold
Alla Weinberg
Design Teams Need Psychological Safety: Here’s How to Create It
2022 • DesignOps Summit 2022
Gold
Kurdin Bazaz
Culture, DIBS & Recruiting
2021 • Design at Scale 2021
Gold
Maria Giudice
Remaking the Making Company: Moving from Product to Experience
2016 • Enterprise UX 2016
Gold
Yalenka Mariën
Designing for Digital Inclusion in the Belgian Government
2021 • Civic Design 2021
Gold
Dave Malouf
The Future of DesignOps
2024 • DesignOps Summit 2020
Gold
Nicole Aleong
Future Orientations to Everyday Life: Futures Anthropology as a Methodology
2024 • Advancing Research 2024
Gold
Audrey Crane
Shadow Design–Where Else is Design Happening in Your Organization? (Videoconference)
2023 • Enterprise Community
Sarah Brooks
Theme Three Intro
2022 • Civic Design 2022
Gold
Josh Clark
Sentient Design: New Design Patterns for New Experiences (3rd of 3 seminars)
2025 • Rosenfeld Community
Sarah Barrett
The "How" of Enterprise Information Architecture
2023 • Enterprise UX 2023
Gold
Rusha Sopariwala
Remote, Together: Craft and Collaboration Across Disciplines, Borders, Time Zones, and a Design Org of 170+
2022 • Design at Scale 2022
Gold
Craig Villamor
Design Systems for Ethical Design (Videoconference)
2023 • Enterprise Community
Jackie Velasquez-Ross
Talent Acquisition and Our Responsibility (Videoconference)
2020 • DesignOps Community
Alla Weinberg
Healing Toxic Stress
2024 • DesignOps 2024
Gold
Darian Davis
Lessons from a Toxic Work Relationship
2024 • Enterprise Experience 2020
Gold

More Videos

Catt Small

"Titles mean different things at different companies, so asking about expectations for levels is critical."

Catt Small Micah Bennett Brian Carr Jessica Harllee

What's Next for ICs: Exploring Staff and Principal Designer Roles (Videoconference)

February 22, 2024

Jennifer Kong

"The current LLM models are great at summarizing, but they’re not so great at capturing exact details."

Jennifer Kong

[Case study] Journeying toward AI-assisted documentation in healthcare

June 5, 2024

Sara Logel

"We’re selectively skeptical – skeptical about some things but not others based on what we want to believe."

Sara Logel

Your Colleagues are Your Users Too

March 29, 2023

Veevi Rosenstein

"At first, product managers often skipped getting customers involved or talked to the same four or five friendly customers repeatedly."

Veevi Rosenstein

Building for Scale: Creating the Zendesk UX Research Practice

January 8, 2024

Michelle Bejian Lotia

"We deliberately use project folders as work-in-progress spaces before migrating insights into the main hub."

Michelle Bejian Lotia Anne-Marie Morell

Rolling Out a Repository: How Zapier Centralizes Insights from Across their Organization

March 28, 2023

Tutti Taygerly

"Share your story, be vulnerable, go first, and reveal what truly matters to you."

Tutti Taygerly

Videconference: How to Work with Difficult People with Tutti Taygerly

June 25, 2020

Patrick Commarford

"We created a simple formula focused on how many people a project impacts and how often they use it."

Patrick Commarford

Design Staffing for Impact

January 8, 2024

Frances Yllana

"AI can help us organize intelligence and content at scale even with limited resources."

Frances Yllana

Theme 2 Intro

September 24, 2024

Mujtaba Hameed

"Draw your ideas, even if you’re a words person; it unlocks new connections and thinking."

Mujtaba Hameed

Frameworks for Excellence: Using Visual Thinking and Communication to Elevate Your Research

March 26, 2024