OpenAI has swiftly rolled back a recently launched ChatGPT feature that let users make their conversations discoverable by search engines like Google. Although the feature was opt‑in—requiring users to select a chat and tick a “Make this chat discoverable” box—reports revealed that thousands of shared conversations, some containing sensitive personal or professional information, were being indexed publicly.
The move came after privacy advocates sounded the alarm over unintended exposure of private dialogues, including discussions about health, relationships, resumes, or client strategies. OpenAI’s Chief Information Security Officer, Dane Stuckey, described the rollout as a “short‑lived experiment” and acknowledged that it introduced too many risks of accidental over sharing. The company said it is actively working with Google and other search providers to de‑index already indexed content and removed the feature from all users within hours.
What happens next? OpenAI plans to refine how shared content is handled—including stronger defaults and clearer user controls. Users are urged to review and delete previously shared links via their ChatGPT Settings → Data Controls → Shared Links dashboard. OpenAI is expected to launch stricter review mechanisms before experimenting with similar public-sharing capabilities in the future.
Photo credit: Unsplash