Incognito conversations

Incognito conversations on claude.ai are temporary chat sessions that are deliberately excluded from your visible chat history and from Claude's persistent memory. Unlike regular chats, what you type in an incognito session does not influence Claude's memory synthesis across future conversations, does not appear in chat search results, and is not used to train Anthropic's AI models. This makes incognito mode useful when you want to explore sensitive topics, work with confidential data, or simply get a clean-slate interaction without any context bleed from previous sessions. When you open an incognito chat, the interface signals the change clearly: a black border frames the conversation window and an 'Incognito chat' label appears in the upper left corner. The session ends when you click the X in the upper right corner, at which point the conversation disappears from your view. However, 'disappears from your view' is not the same as 'immediately and permanently deleted.' Anthropic retains incognito chats on the backend for at least 30 days for safety purposes, and Enterprise organizations may have longer custom retention periods. Incognito mode is not a tool for bypassing Anthropic's safety systems or for hiding activity from your employer. If a safety classifier flags content in an incognito chat, that content may still be reviewed for trust and safety purposes. On Team and Enterprise plans, account Owners can access incognito chats through organizational data exports and the Compliance API, meaning incognito provides user-level privacy from persistent memory and model training — not organizational-level anonymity.