By first using the AI chatbot in one tab and later activating it in another tab, the document title of the previous tab would leak into the chat prompt, revealing page titles or URLs that could aid an attacker in identifying sensitive or internal resources. This vulnerability results in the unintended disclosure of confidential information to a user who manipulates the AI chat feature across tabs, representing a moderate information‑disclosure risk.

Mozilla: Firefox. Versions up to and including Firefox 136 are vulnerable; the issue was fixed in Firefox 137.

The CVSS score of 5.3 indicates a moderate overall risk, while the EPSS score of less than 1% suggests a very low exploitation probability at the time of assessment. The vulnerability is not listed in CISA’s KEV catalog. An attacker would need to be a local user who can open multiple tabs with the AI chatbot active, inferring that the attack vector is user‑initiated browser activity rather than remote exploitation.