started · updated
Anthropic faces privacy crisis as Claude chat logs appear in Google searches
A privacy crisis has emerged regarding Anthropic’s AI assistant, Claude, after user chat logs and generated content were discovered in Google search results. The issue was brought to light by Reddit users who utilized specific search commands to find publicly indexed chat-sharing links.
While Claude’s sharing feature allows users to create unique links to share dialogues with specific individuals, these links were captured by search engine crawlers and added to public directories. The exposed data includes highly sensitive information, such as personal health records, clinical research results, confidential corporate correspondence, and employee performance evaluations.