Ad Inserter EUVD-2026-53865

Exploitation requires only network access to a WordPress installation running Ad Inserter 2.8.16 or earlier. The CVSS vector (AV:N/AC:L/PR:N/UI:N) confirms no authentication, no elevated privileges, no special plugin configuration, and no victim interaction are needed - the vulnerable ai_ajax endpoint is accessible in the plugin's default installed state. The primary limiting factor on impact is data sensitivity: only ad block contents explicitly configured by an administrator as 'administrator-only visibility' are exposed; installations where no such restriction is applied gain no confidentiality loss even if the endpoint is invoked. Exploitation does not grant write access, escalate privileges, or affect availability.

The CVSS 5.3 Medium score accurately reflects the limited blast radius: attack complexity is low, no authentication or user interaction is required (consistent with the AV:N/AC:L/PR:N/UI:N vector), but the confidentiality impact is bounded to C:L with no integrity or availability consequences. In practice, the exposed data is ad block configuration content - not credentials, PII, or system data - which narrows operational risk considerably. EPSS data was not provided in the source intelligence; KEV status is not confirmed, and no active exploitation evidence exists in this dataset. The real-world priority depends on whether restricted ad blocks contain sensitive business logic or proprietary configurations that an administrator explicitly chose to conceal; for most deployments this is low-sensitivity data. Two plugin changesets suggest the vendor has already responded, which further reduces urgency for sites running auto-update or actively maintained WordPress installations.

An unauthenticated attacker sends a crafted HTTP POST request to wp-admin/admin-ajax.php on any vulnerable WordPress site, specifying the ai_ajax action parameter. Because the handler performs no capability check, the server returns the contents of administrator-restricted ad blocks without requiring credentials. No public proof-of-concept has been identified at time of analysis, but the attack is entirely self-contained and requires only knowledge of the vulnerable action name.