LLMs are getting better at qualitative analysis tasks. But when errors do show up, they're often important. And they’re now almost impossible to catch by just eyeballing the output.

This should be a wake-up call for all of us using AI as a research tool. We can’t trust that our familiarity with the data is enough to protect us from making mistakes.

If you could get an inside look at how AI-moderated interviews work under the hood, what would you want to know?

"AI analysis ≠ qualitative analysis" is one of most misunderstood things I teach about AI for UX research. I'm not saying that AI analysis is *worse quality* than human qual analysis. I'm saying that if you look under the hood, they're fundamentally different processes.

Our workshop tools are designed to give you a transparent view inside the black box of commercial research tools. With our new tool, you get direct access to the prompts that guide the AI moderator, so you can see for yourself how changes impact your results.

The prompts and infrastructure behind commercial AI research tools are a closely guarded secret. But if you can’t know what’s happening with your research data once it goes into the tool, you can’t know whether that process can be trusted to yield reliable results. I asked if Condens could do something different: share a basic account of their specific AI safeguards.

I’ve been updating my AI for UX Researchers workshop with new content for 2026. Here’s a sneak peak of the outline.

Teaching general principles only gets you so far when you’re working with tools that change as rapidly as AI. I see my workshops as less about teaching static best practices, and more about teaching researchers how to think about AI–and that starts with learning how to experiment with the latest models and assess the quality of their output.

Verifying qualitative research analysis falls into Andrej Karpathy’s category of “lagging” AI tasks, depending, as it does on “real-world knowledge…context, and common sense.” There’s no definitive “correct answer” to train a computer by. So what can we use AI for in product research?

We tested Google’s Gemini 2.5 Flash (used in NotebookLM) on qualitative analysis by running the same prompt multiple times.  It improved on GPT-4o and Claude Sonnet 3.5 at quote accuracy but still struggled with relevance and meaning.

NotebookLM is useful, but it’s still just LLMs deep down. There’s no magic technology that makes it immune to the problems of other AI systems.

Seen last week on social media: a market researcher trying to save their job after reporting a nonsense, LLM-generated analysis of survey data. It’s a common mistake, and there’s not much training available for researchers on how to account for differences between generative AI and traditional research tools.

The reality is that many researchers are now expected to use LLM-based tools in their work, despite their shortcomings. And it can be hard to push back on mandates from above, or competition from peers who don’t understand the risks. Are there ways we can integrate LLMs into research while still being responsible professionals about it?

When we use AI tools in product design, strategy, and user research, we need to keep something important in mind: the AI is not “understanding” our requests, and even with safeguards, it may not catch when it’s making mistakes.