This vulnerability in VMware Spring AI arises from improper access control (CWE-284) where an attacker can manipulate the conversationId parameter to bypass conversation isolation. This enables unauthorized access to sensitive memory containing other users' chat histories, including secrets and credentials. The issue specifically impacts applications that use VectorStoreChatMemoryAdvisor and accept user input directly as conversationId. The CVSS 3.1 base score is 5.9, reflecting a medium severity with network attack vector, high attack complexity, no privileges required, no user interaction, and high confidentiality impact but no integrity or availability impact. No official patch or remediation guidance is currently available from the vendor.

Successful exploitation allows an attacker to access sensitive memory from other users' chat histories, potentially exposing secrets and credentials. This compromises confidentiality but does not affect integrity or availability. The vulnerability requires network access and has a high attack complexity, meaning exploitation is not trivial. No known exploits are currently reported in the wild.

Patch status is not yet confirmed — check the vendor advisory for current remediation guidance. Until an official fix is released, applications should avoid passing user-supplied input directly as conversationId to VectorStoreChatMemoryAdvisor or implement strict validation and sanitization of this input to prevent injection of filter logic. Monitor vendor communications for updates on patches or official mitigations.