Anthropic Debuts Interviewer Tool to Understand How People Really Use AI

Anthropic has rolled out a new research tool, Anthropic Interviewer, designed to track how people use artificial intelligence and how they feel about it. 

Starting 4 December, select users of Claude.ai will see a pop-up inviting them to participate, marking a fresh phase in the company’s large-scale study of human-AI interaction. 

The tool runs fully automated interviews powered by Claude, producing qualitative data at scale that can be sifted by human researchers.

In its first pilot, Anthropic interviewed 1,250 professionals, including workers across the general workforce, creatives and scientists, and has published anonymized transcripts for public analysis. 

Early findings show widespread optimism about AI’s time-saving potential: 86% of general-workforce respondents said AI “saves them time.” 

Most professionals prefer AI as a collaborator rather than a replacement and imagine future roles centered on supervising AI systems.

Among creative professionals, 97% said AI improved productivity, though many reported social stigma or peer pressure to use it just to remain competitive. 

Scientists were more cautious: while most use AI tools for tasks like literature review or drafting manuscripts, many said current systems are not reliable enough for core research duties such as hypothesis generation, yet 91% expressed interest in deeper AI support in the future. 

Anthropic said the new tool is a step toward building a richer, research-backed understanding of how AI fits into real work lives, and could support broader collaborations with creatives, scientists, teachers and everyday users.