Log in or create a free Rosenverse account to watch this video.
Log in Create free account100s of community videos are available to free members. Conference talks are generally available to Gold members.
AI-Assisted Qualitative Analysis: What to Automate and What to Own
Summary
Qualitative analysis is about more than summarizing common trends in your data. It’s about prioritizing and explaining what’s important, and that requires getting familiar with your data and thinking hard about what you're seeing. AI can support that process, but it can also short-circuit the thinking you need to do if you're not careful. In this session, we'll look at what qualitative analysis actually involves, where AI genuinely helps, and where skipping the process costs you. You'll come away with a practical sense of how to use AI as an accelerant and thought partner while maintaining control of the work.
Key Insights
-
•
Qualitative analysis involves moving beyond patterns to generating meaningful explanations and insights.
-
•
AI works best when humans fully understand the research process and guide AI step-by-step.
-
•
Structured data can be heavily automated by AI, reducing manual coding efforts.
-
•
Messy, rich qualitative data requires strong human oversight and involvement to capture unexpected insights.
-
•
Humans bring subjective intuition and surprise to analysis, which AI currently cannot replicate.
-
•
Keeping humans in the loop means supervising major decisions, reviewing AI outputs, and continually refining codes.
-
•
The thematic analysis process benefits from breaking work into discrete steps for AI to perform under human supervision.
-
•
Cognitive offloading risks dulling critical qualitative research skills if AI is overused.
-
•
Adversarial AI agents can be employed to challenge and verify AI-generated findings before human review.
-
•
AI tools can assist in report writing by handling repetitive tasks like tallying but struggle to replicate human narrative tone and excitement.
Notable Quotes
"If you don't understand the process, you're not going to get very far with using AI for analysis."
"AI is a tool. It can be as powerful as the person who is wielding that tool."
"Qualitative analysis isn’t just to call out patterns but to explain why they’re there and make them meaningful for your audience."
"Automate the hell out of structured data, but stay involved in messy, rich data."
"Think of AI as an extra colleague: it does the first pass, you do the second pass."
"Coding is more important than the output; it forces you to be familiar with and think deeply about your data."
"Use AI to remove boring, menial tasks but preserve the manual, difficult hard work for yourself."
"Slowing down on messy data is worth it because you could miss something incredibly important that changes the trajectory of the product."
"Adversarial verification with a second AI agent trying to poke holes in findings mimics how human researchers challenge each other."
"AI doesn’t feel surprise or curiosity, but humans do, and that’s a unique strength in analysis."
Or choose a question:
More Videos
"If given a choice in my career and my personal life, I will choose process over ambiguity every single time."
Amy ThibodeauOpening Keynote: Process and Ambiguity
October 23, 2019
"Managers were asking for user stories but hadn’t done any analysis or design — they didn’t understand the scope."
Carl TurnerYou Can Do This: Understand and Solve Organizational Problems to Jumpstart a Dead Project
March 28, 2023
"It’s worth explaining your new communication approach to stakeholders to save their time and foster discussion."
Bruce GillespieLearning from journalism: Balancing impactful communication with compassionate storytelling
March 13, 2025
"Design producers love their creative tools and want reporting to be automated, not manual."
Aurobinda Pradhan Shashank DeshpandeIntroduction to Collaborative DesignOps using Cubyts
September 9, 2022
"Drag and drop ease, and collaborative affinity diagramming in the tool, were unexpected but very welcome advantages."
Taylor Jennings Joe Nelson Alex KnollRepository Retrospective: Learnings from Introducing a Central Place for UX Research
March 9, 2022
"You have to keep the injection sites moving because of scar tissue, adding to the body hacking complexity."
Daniel J. RosenbergDigital Medicine Design
September 26, 2019
"Culture needs to accept failure as part of learning so we can improve rather than assign blame."
Jack MoffettUX Metrics That Matter and The Future of our Design at Scale Conference: A Community Conversation
September 22, 2022
"I thought I could skip the architecture step, but I ended up with spaghetti code that broke constantly."
Christian CrumlishThe Pygmalion Effect: In Which a Vibe Coding Experiment Becomes a Million Lines…
August 14, 2025
"The more participatory the practices, the deeper and more intentional the engagement needs to be."
Alexia Cohen Adriane AckermanIncreasing Health Equity and Improving the Service Experience for Under-Served Latine Communities in Arizona
December 4, 2024
