A recently patched sandbox vulnerability in ServiceNow, which could allow for remote code execution (RCE), is now being exploited in the wild, as highlighted by threat intelligence firm Defused. Their report indicated that attackers are now employing new tactics that diverge from earlier methods documented in a proof of concept (PoC) by Searchlight Cyber, adapting to the protections added by ServiceNow.
Defused CEO Simo Kohonen remarked that the evolving tactics show a marked increase in the variety of attack methods, saying, “We are seeing a lot of variations, much more so than a year ago, for the same vulnerability.” So far, however, this particular exploit has only been observed in action by one actor.
In response to these developments, ServiceNow stated that, based on their investigations, they haven’t observed any exploits directly impacting their hosted instances. They emphasized that updates and patches have been released to mitigate this issue and encouraged both self-hosted and ServiceNow-hosted customers to apply these patches promptly.
A Critical Vulnerability in Sandbox Architecture
Analysts have expressed concern over the implications of this vulnerability, emphasizing its potential to compromise a key layer of security that many IT and security teams have relied upon for years. According to Frank Dickson, group VP for security at IDC, the vulnerability allows attackers to bypass ServiceNow's scripting sandbox entirely, complicating defenses that are typically based on documented attack methodologies.
“A compromise that starts in the cloud tenant can end up inside the corporate network, turning a SaaS incident into an on-premises one,” he noted, adding that the data housed within ServiceNow often includes sensitive HR records and asset management information. Attackers exploiting the vulnerability could gain critical insights into internal incident tracking.
This situation indicates that organizations may need to rethink their patching strategies. Dickson pointed out that while enterprises often entrust patch management to vendors, they still carry risks associated with those platforms, particularly as these services have connections to on-premises systems. “Control sits with the vendor, liability sits with the enterprise,” he explained, urging CISOs to reconsider how they perceive risks associated with these vital platforms.
Additionally, Noah Kenney, a principal consultant at Digital 520, emphasized the troubling nature of the sandbox escape aspect. He argued that the failure of the containment layer, designed to safely run untrusted AI-driven code, contradicts assurances provided by many vendors that their sandboxes ensure safety. This incident should prompt CISOs to reassess all features running behind similar security walls.
AI's Impact on Security Dynamics
The incorporation of AI technologies into enterprise systems is shifting traditional IT and security paradigms. Kenney noted, “Enterprises are bolting AI onto their most privileged systems of record faster than anyone is updating the threat models for those systems.” The potential impact of this integration is concerning, as the AI layer often becomes a vulnerable vector for attacks.
Aman Mahapatra, chief strategy officer at Tribeca Softtech, echoed these sentiments, stating that new vulnerabilities now provide attackers with footholds that extend far beyond traditional exploitation. “The blast radius of a ServiceNow compromise in 2026 is meaningfully larger than the same compromise would have been in 2023,” he warned, indicating that security measures are lagging behind the rapid evolution of these threats.
While Kohonen acknowledged the concerns raised about sandbox security, he stressed a crucial point: many CISOs have recognized the limitations of relying purely on sandbox environments for security. “Nothing is foolproof, and having a sandbox is better than not having one. But believing a sandbox removes all risk is incredibly naive.”
This ongoing situation serves as a stark reminder for organizations to not only apply necessary patches but also to constantly evaluate their security infrastructures. As the threat landscape evolves, businesses must adapt their defenses to include a more comprehensive understanding of how integrated technologies, particularly AI, alter their risk profiles.
