CVE-2026-14549: Ray Ray Enterprise Translation WP

CISO Take

The Ray Enterprise Translation WordPress plugin (versions through 1.7.3) fails to perform any capability or nonce check on one of its AJAX actions, letting any authenticated user — even a low-privilege Subscriber — add or delete the site's configured languages. Despite being tagged in our feed under the ml_ops AI category, this is a generic WordPress content-management plugin unrelated to the "ray" distributed-compute/ML-ops package tracked in our AI package inventory; this looks like a name-collision artifact rather than a genuine AI/ML supply-chain finding, and it carries essentially no blast radius for AI pipelines. Exploitation is trivial (CVSS 4.3, PR:L, no confidentiality or availability impact, integrity impact limited to language config), there is no public exploit, no Nuclei template, and CISA's SSVC decision is TRACK_STAR — the lowest-priority tracking tier. The EPSS score of 0.152% confirms very low real-world exploitation probability. Action: monitor for a vendor patch beyond 1.7.3 via the WPScan advisory, restrict Subscriber-role registration if not required, and flag this record for AI-category reclassification since it does not represent an AI/ML risk despite its current tagging.

Sources: NVD EPSS wpscan.com

What is the risk?

Low risk overall. The vulnerability is a broken access control / missing nonce check (CVSS 3.1: 4.3, medium) that any authenticated Subscriber-level WordPress user can trigger with a forged AJAX request. There is no confidentiality or availability impact — only limited integrity impact (unauthorized add/delete of configured languages). No public exploit code, no Nuclei scanning template, not in CISA KEV, and CISA's own SSVC decision is TRACK_STAR (lowest-priority, track-only). EPSS raw probability is 0.152%, reinforcing that near-term mass exploitation is unlikely. Critically, this CVE has no genuine bearing on AI/ML systems — it affects a WordPress translation plugin whose name happens to collide with the tracked "ray" ml_ops package, and should not be treated as an AI supply-chain threat.

How does the attack unfold?

Initial Access

Attacker obtains or registers a low-privilege Subscriber account on a WordPress site running the vulnerable plugin.

Exploitation

Attacker sends a forged AJAX request to admin-ajax.php targeting the unprotected language-management action, bypassing missing capability and nonce checks.

Impact

Site's configured languages are added or deleted without authorization, disrupting content localization and site integrity.

Initial Access

Attacker obtains or registers a low-privilege Subscriber account on a WordPress site running the vulnerable plugin.

Exploitation

Attacker sends a forged AJAX request to admin-ajax.php targeting the unprotected language-management action, bypassing missing capability and nonce checks.

Impact

Site's configured languages are added or deleted without authorization, disrupting content localization and site integrity.

What systems are affected?

Package	Ecosystem	Vulnerable Range	Patched
Ray	pip	—	No patch
43.9K OpenSSF 5.8 635 dependents Pushed 5d ago 68% patched ~145d to patch Full package profile →

Do you use Ray? You're affected.

How severe is it?

CVSS 3.1

4.3 / 10

EPSS

0.3%

chance of exploitation in 30 days

Higher than 15% of all CVEs

Source: EPSS v3 — FIRST.org

Exploitation Status

Exploit Available

Exploitation: MEDIUM

Sophistication

Trivial

Exploitation Confidence

medium

○ CISA SSVC: Public PoC

○ Public PoC indexed (trickest/cve)

Composite signal derived from CISA KEV, VulnCheck KEV, CISA SSVC, EPSS, Metasploit, Exploit-DB, trickest/cve, Nuclei templates, and inthewild.io exploitation reports.

What is the attack surface?

AV Network

AC Low

PR Low

UI None

S Unchanged

C None

I Low

A None

What should I do?

1 step

1) Track the vendor/WPScan advisory (https://wpscan.com/vulnerability/3dd9bf3b-7fe4-4a58-ad05-1fc0bfb85413/) for a fixed release beyond 1.7.3 and update as soon as one is published. 2) In the interim, disable open Subscriber self-registration on affected WordPress sites, or restrict which roles can reach admin-ajax.php actions via a security plugin (e.g., enforce capability checks with a WAF/plugin like Wordfence). 3) Monitor server logs for anomalous admin-ajax.php POST requests referencing the plugin's language-management action from non-admin accounts. 4) Since this CVE has no material AI/ML relevance, flag the record internally for ai_category correction (currently mistagged ml_ops due to name overlap with the unrelated "ray" pip package) so it doesn't skew AI threat metrics or trigger unnecessary customer-facing alerts.

What does CISA's SSVC say?

Decision Track*

Exploitation poc

Automatable No

Technical Impact partial

Source: CISA Vulnrichment (SSVC v2.0). Decision based on the CISA Coordinator decision tree.

How is it classified?

Auth Bypass Plugin

Which compliance frameworks are affected?

This CVE is relevant to:

NIST AI RMF

MAP-3.1 - Third-party AI/ML component risk tracking

Frequently Asked Questions

What is CVE-2026-14549?

The Ray Enterprise Translation WordPress plugin (versions through 1.7.3) fails to perform any capability or nonce check on one of its AJAX actions, letting any authenticated user — even a low-privilege Subscriber — add or delete the site's configured languages. Despite being tagged in our feed under the ml_ops AI category, this is a generic WordPress content-management plugin unrelated to the "ray" distributed-compute/ML-ops package tracked in our AI package inventory; this looks like a name-collision artifact rather than a genuine AI/ML supply-chain finding, and it carries essentially no blast radius for AI pipelines. Exploitation is trivial (CVSS 4.3, PR:L, no confidentiality or availability impact, integrity impact limited to language config), there is no public exploit, no Nuclei template, and CISA's SSVC decision is TRACK_STAR — the lowest-priority tracking tier. The EPSS score of 0.152% confirms very low real-world exploitation probability. Action: monitor for a vendor patch beyond 1.7.3 via the WPScan advisory, restrict Subscriber-role registration if not required, and flag this record for AI-category reclassification since it does not represent an AI/ML risk despite its current tagging.

Is CVE-2026-14549 actively exploited?

Proof-of-concept exploit code is publicly available for CVE-2026-14549, increasing the risk of exploitation.

How to fix CVE-2026-14549?

1) Track the vendor/WPScan advisory (https://wpscan.com/vulnerability/3dd9bf3b-7fe4-4a58-ad05-1fc0bfb85413/) for a fixed release beyond 1.7.3 and update as soon as one is published. 2) In the interim, disable open Subscriber self-registration on affected WordPress sites, or restrict which roles can reach admin-ajax.php actions via a security plugin (e.g., enforce capability checks with a WAF/plugin like Wordfence). 3) Monitor server logs for anomalous admin-ajax.php POST requests referencing the plugin's language-management action from non-admin accounts. 4) Since this CVE has no material AI/ML relevance, flag the record internally for ai_category correction (currently mistagged ml_ops due to name overlap with the unrelated "ray" pip package) so it doesn't skew AI threat metrics or trigger unnecessary customer-facing alerts.

What is the CVSS score for CVE-2026-14549?

CVE-2026-14549 has a CVSS v3.1 base score of 4.3 (MEDIUM). The EPSS exploitation probability is 0.25%.

What is the AI security impact?

Compliance Controls Affected

NIST AI RMF: MAP-3.1

What are the technical details?

Original Advisory

The Ray Enterprise Translation WordPress plugin through 1.7.3 does not perform any capability or nonce checks on one of its AJAX actions, allowing any authenticated user, including Subscribers, to add or delete the site's configured languages.

Exploitation Scenario

An attacker obtains or registers a low-privilege Subscriber account on a WordPress site running Ray Enterprise Translation ≤1.7.3 (via open registration or a compromised low-tier account). Because the plugin's AJAX handler performs no capability or nonce validation, the attacker sends a direct POST request to admin-ajax.php with the relevant action parameter — no UI interaction or elevated privileges required — and adds bogus languages or deletes the site's configured languages, disrupting the site's localization/content structure without needing admin access.