Anthropic has turned its own AI, Claude, into a large-scale research interviewer, capable of running thousands of structured, 10–15 minute qualitative interviews without a human moderator in the room. In its first major outing, the tool spoke to 1,250 professionals about how AI is changing their work, creating one of the richest public datasets yet on day-to-day AI use.
The system runs the full pipeline: it helps design the research rubric, conducts adaptive conversations, and then clusters responses into themes for human analysts to interpret. This means open-ended, in-depth interviews —traditionally slow and expensive—can now be done at a scale previously reserved for clickstream analytics and multiple-choice surveys.
The first findings are telling: 86% of professionals say AI saves them time and 65% are happy with its role, yet 69% still feel a social stigma in openly using AI tools and 55% worry about what this means for their future. Creatives report hiding their AI use and fearing displacement, while many scientists see AI as a promising research partner that they do not fully trust yet.
By releasing all 1,250 interview transcripts publicly and planning repeat waves, Anthropic is not just testing a product; it is redefining how organisations can listen to users and workers in the AI era. If this model spreads, market research, policy consultations, employee feedback, and even journalism could move from sample surveys to always-on, deeply documented conversations at population scale.
INTERVIEWS AT SCALE, INSIGHT AT SPEED.