Proton has launched AI Paper Trail, a free diagnostic utility designed to audit exported user conversation histories from prominent artificial intelligence platforms and calculate individual exposure risks.
The software inspects user data archives exported from AI models such as ChatGPT and Claude, systematically parsing previous interactions to generate a comprehensive personal privacy report. According to Proton, uploaded user datasets undergo immediate deletion subsequent to analysis and are not retained on Lumo server infrastructure.
Enterprise and consumer interactions with AI assistants gradually aggregate vast repositories of sensitive metadata. While individual prompts often appear innocuous in isolation, cumulative histories recorded over extended periods frequently divulge intimate details concerning professional endeavors, interpersonal relationships, daily routines, consumer purchasing habits, and upcoming travel itineraries.
During practical evaluations utilizing personal ChatGPT conversation archives on the free tier, the diagnostic framework processed the 200 most recent user prompts. The assessment yielded an AI Exposure Score of 58 out of 100, categorizing the profile under the ‘Leaving receipts’ designation while identifying 47 distinct data points and estimating an advertising value of $185.
The utility categorized evaluated risks across multiple verticals, highlighting location tracking, technology utilization, and personal interests as primary vectors of data leakage. Proton emphasizes that while the generated reports rely heavily on algorithmic inferences rather than absolute empirical facts, the output demonstrates how rapidly seemingly disconnected user queries coalesce into detailed behavioral profiles.