Executive brief
Open WebUI is a user-friendly interface for interacting with AI models. The vulnerability allows administrators to view private chat conversations of other administrators by modifying a web parameter, potentially exposing sensitive or confidential discussions. Although originally reported as a medium-severity issue, the advisory was subsequently withdrawn because further analysis determined it does not represent a valid security vulnerability.
Technical details
This is an authorization bypass vulnerability (CWE-639) in Open WebUI's frontend admin dashboard. The vulnerability occurs in the chat-viewing functionality where administrators should only see chats from non-admin users. However, by directly modifying the user_id parameter in requests, an authenticated admin can craft requests to access chat records of other admin or owner accounts. The attack vector is network-based, requires high privileges (admin authentication), and has no integrity or availability impact—only confidentiality loss. Attack preconditions include valid admin credentials and network access to the admin dashboard. The advisory was withdrawn on September 2, 2026, indicating that upon further review, the finding does not meet the criteria for a valid security vulnerability, though the original report referenced affected versions through 0.3.8 with no patched version identified at that time.
Affected products
- Open WebUI open-webui <= 0.3.8
Timeline
- 2025-03-20: disclosed: Advisory published to GitHub Advisory Database
- 2026-09-02: other: Advisory withdrawn as not describing a valid vulnerability