Agentic LLMs as Powerful Deanonymizers: Re-identification of Participants in the Anthropic Interviewer Dataset

TL;DR

This paper demonstrates that modern agentic large language models can effectively re-identify participants in qualitative interview datasets by linking interview details to publicly available information, raising privacy concerns.

Contribution

It reveals that off-the-shelf LLM agents can perform re-identification attacks easily, highlighting privacy risks in releasing qualitative datasets.

Findings

LLMs can link interviews to specific scientific works and authors.

Re-identification can be achieved with minimal prompts and effort.

Existing safeguards are insufficient against LLM-based re-identification.

Abstract

On December 4, 2025, Anthropic released Anthropic Interviewer, an AI tool for running qualitative interviews at scale, along with a public dataset of 1,250 interviews with professionals, including 125 scientists, about their use of AI for research. Focusing on the scientist subset, I show that widely available LLMs with web search and agentic capabilities can link six out of twenty-four interviews to specific scientific works, recovering associated authors and, in some cases, uniquely identifying the interviewees. My contribution is to show that modern LLM-based agents make such re-identification attacks easy and low-effort: off-the-shelf tools can, with a few natural-language prompts, search the web, cross-reference details, and propose likely matches, effectively lowering the technical barrier. Existing safeguards can be bypassed by breaking down the re-identification into benign tasks. I outline the attack at a high level, discuss implications for releasing rich qualitative data in the age of LLM agents, and propose mitigation recommendations and open problems. I have notified Anthropic of my findings.