This vulnerability (CWE-639) in Kibana permits an attacker with some privileges to bypass authorization controls by exploiting user-controlled keys. It enables unauthorized access to inference outputs of trained models in different spaces, which the attacker is not authorized to list, read, or use. Additionally, it allows stopping or updating active trained model deployments in other spaces, potentially altering allocated resources. The vulnerability is identified in Kibana versions 8.19.0, 9.0.0, and 9.5.0. The CVSS 3.1 score is 7.1 (high severity), reflecting network attack vector, low attack complexity, low privileges required, no user interaction, unchanged scope, high confidentiality impact, no integrity impact, and low availability impact.

Successful exploitation leads to unauthorized disclosure of trained model inference outputs from other spaces, exposing sensitive model behavior. It also permits unauthorized stopping or updating of active model deployments in other spaces, which could disrupt model availability or alter resource allocation. There is no reported impact on integrity, but confidentiality is highly affected and availability impact is low.

Patch status is not yet confirmed — check the vendor advisory for current remediation guidance. No official fix or temporary workaround is indicated in the provided data. Users should monitor Elastic's advisories for updates and apply patches once available. Until then, restrict user privileges and monitor access to trained model functionalities to minimize risk.