Categories: Cyber Security News

Critical ServiceNow Vulnerability Enables Privilege Escalation via Unauthenticated User Impersonation

A critical privilege-escalation vulnerability has been discovered in ServiceNow’s AI Platform, posing substantial security risks to enterprises worldwide.

The flaw, tracked as CVE-2025-12420, allows unauthenticated attackers to impersonate legitimate users and execute unauthorized operations leveraging the compromised account’s permissions, a dangerous capability that could grant threat actors unfettered access to sensitive business functions.

The vulnerability was identified by AppOmni, a leading SaaS security firm, which responsibly reported it to ServiceNow in October 2025 via coordinated disclosure channels.

ServiceNow responded with commendable speed, deploying security patches to its hosted instances by October 30, 2025, while simultaneously providing updates to partners and customers operating self-hosted deployments.

Nevertheless, the company remains vigilant, noting no confirmed active exploitation in the wild despite heightened risk following public disclosure.

Two critical applications require immediate attention. The Now Assist AI Agents application demands patching to version 5.1.18 or later, or alternatively, version 5.2.19 or later.

The Virtual Agent API also requires an update to version 3.15.2 or later, or to version 4.0.4 or later.

ServiceNow has emphasized the urgency of these updates in its knowledge base, which provides comprehensive security maintenance guidance for both hosted and self-hosted environments.

Organizations utilizing ServiceNow’s AI capabilities for business-critical operations should prioritize patching efforts immediately.

The privilege escalation vector poses a particularly acute threat, as attackers could pivot from initial compromise to accessing resources with the impersonated user’s role and permissions.

This creates cascading risks across interconnected systems and sensitive data repositories.

Security teams should verify their current ServiceNow deployments against the affected versions and initiate patch deployment without delay.

Given the vulnerability’s unauthenticated attack vector and high severity, this should be a top priority in organizational remediation workflows.

ServiceNow’s rapid response and patch availability reduce the window for exploitation, but proactive patching remains essential to eliminate attack vectors before sophisticated threat actors can develop weaponized exploits.

Field Value
CVE ID CVE-2025-12420
Vulnerability Type Privilege Escalation
Affected Product ServiceNow AI Platform

Follow us on Google News , LinkedIn and X to Get More Instant UpdatesSet Cyberpress as a Preferred Source in Google.

The post Critical ServiceNow Vulnerability Enables Privilege Escalation via Unauthenticated User Impersonation appeared first on Cyber Security News.

rssfeeds-admin

Recent Posts

Rhythm Heaven Groove for Nintendo Switch Is Up for Preorder

Originally announced at a Nintendo Direct all the way back in March 2025, Rhythm Heaven…

38 minutes ago

How Can We Fix the Ongoing Ascended Heroes Price Crisis in Pokémon TCG?

If you, like me, are looking to complete your Pokémon TCG binder, we may have…

39 minutes ago

Project Hail Mary Rocky Actor Could Get Oscar Nomination

Audiences are really loving Project Hail Mary’s Rocky, the sentient rock-based alien lifeform who charms…

39 minutes ago

Firefox 150 Released With Fixes for Multiple Code Execution Vulnerabilities

Mozilla has released Firefox 150, addressing 41 security vulnerabilities, including multiple high-severity flaws that could…

1 hour ago

Critical Spring Authorization Server Flaw Enables XSS, Privilege Escalation, and SSRF

A critical security vulnerability, tracked as CVE-2026-22752, has been discovered in Spring Security Authorization Server,…

1 hour ago

Fake Google Ads Used To Steal Seed Phrases and Drain Crypto Wallets

Cybersecurity organization SEAL (Security Alliance) has issued a critical warning about a sustained and escalating…

1 hour ago

This website uses cookies.