Where is AI letting us explore user data in ways that are entirely new?

Latent Scope by Ian Johnson

We talk a lot about whether AI can replace traditional user research functions. But where is AI letting us explore user data in ways that are entirely new?

That’s the question asked by Ian Johnson and Patrick Boehler, in two separate case studies from the very different lenses of data visualization and journalism. I’ve been working with them over the past few months to develop their talks for Designing with AI 2025.

Ian has developed Latent Scope, a tool that allows you to visualize relationships in unstructured data (think survey free response, or even images), using the same kind of math as LLMs. This means you can discover meaningful clusters in large datasets that might never emerge from traditional qualitative analysis. It doesn’t replace user research; instead, it introduces an entirely new sandbox for researchers to play with (and in his case study, Ian explains exactly how to do it).

Patrick approaches data exploration from an entirely different angle, asking how we can quickly generate trustworthy insights from disparate data sources, like market research or white papers. AI can aid in sourcing relevant research from a repository, but it can’t assess source credibility on its own. So Patrick’s team at Gazzetta constructed a confidence rating system that gets applied to every item in their repository, and propagated through their AI-generated insights system.

As a research practitioner, this is the kind of AI work that excites me, because it looks forward instead of backwards, introducing new capabilities rather than (poorly?) automating what we do already.

Every researcher dabbling with AI for their research repositories needs strategies for handling confidence levels, like Patrick is using at Gazzetta.

Every researcher can benefit from being able to discover new types of patterns in unstructured data (not just for text, but also for media), like Ian is demonstrating with Latent Scope.

You can see Ian’s and Patrick’s case studies at Designing with AI in June. I'd love to see you there!