CVE-2026-89278: GPTranslate <= 2.34.6 - Unauthenticated Sensitive Information Exposure in Public Frontend Inline Script
The GPTranslate – Multilingual AI Translation Agent for WordPress: Translate Your Site with AI plugin for WordPress is vulnerable to Sensitive Information Exposure in all versions up to, and including, 2.34.6 via the enqueuefrontendscripts. This makes it possible for unauthenticated attackers to extract the plaintext third-party AI provider API key (OpenAI, DeepL, xAI/Grok, Gemini, Claude, or Google Cloud Translation) — a credential granting billed account access — by fetching any public page and applying the inverse transformation bundled in the plugin's own public JavaScript asset. This exposure affects the default configuration (gpt-3.5-turbo in client mode) and all supported non-DeepSeek providers; only deepseek- models and gpt- models configured in server-proxy mode correctly suppress key emission.
Affected Software
Remediation
Recommended actions to resolve this vulnerability, in priority order.
- Upgrade
Upgrade
GPTranslate – Multilingual AI Translation Agent for WordPress (WordPress plugin)to a version that resolves this vulnerability.Fixed in 2.34.6 - Compensating control
Until fixed, configure the plugin to use only deepseek-* models or gpt-* models in server-proxy mode so key emission is suppressed (plaintext API keys are only correctly suppressed for those configurations).
- Operational
If the public frontend exposed plaintext third-party AI provider API keys (OpenAI, DeepL, xAI/Grok, Gemini, Claude, or Google Cloud Translation), rotate/revoke those keys because they may have been extracted by unauthenticated attackers.
Event History
Frequently Asked Questions
Which deployments are exposed to unauthenticated key retrieval?
The default gpt-3.5-turbo client-mode configuration is affected, as are configurations using supported non-DeepSeek providers. Deepseek-* models and gpt-* models configured in server-proxy mode suppress API-key emission.
What does an attacker need to retrieve the exposed credential?
No authentication or user interaction is required. An attacker can fetch any public page and use the inverse transformation contained in the plugin's public JavaScript asset to recover the plaintext provider API key.
What is the practical impact of a recovered key?
The exposed key can grant access to the associated third-party AI provider account, including billed account usage. The affected providers include OpenAI, DeepL, xAI/Grok, Gemini, Claude, and Google Cloud Translation.
What mitigation is indicated if an update cannot be applied immediately?
Configure gpt-* models to use server-proxy mode, or use a deepseek-* model, because those configurations suppress key emission. Any key that may have been exposed should be treated as compromised and replaced.