CISA Flags Actively Exploited Ray Flaw That Can Trigger Browser-Based RCE

Severity
High

Detail

The U.S. Cybersecurity and Infrastructure Security Agency (CISA) has added a critical vulnerability affecting Ray, an open-source Python framework used for distributed AI and machine learning workloads, to its Known Exploited Vulnerabilities (KEV) catalog after confirming active exploitation.

The vulnerability, tracked as CVE-2025-62593 with a CVSS score of 9.4, can allow remote code execution through a web browser using a DNS rebinding attack. The issue is particularly concerning for developers running Ray locally or within private corporate environments.

How?

The vulnerability exists because Ray exposes certain API endpoints without authentication. An attacker can abuse these endpoints through a DNS rebinding attack, causing a victim’s web browser to send malicious requests to a vulnerable Ray instance.

The attack can be triggered when a victim visits a malicious website or interacts with malicious content while running an affected Ray environment. The browser can act as a proxy to reach Ray services that may otherwise be inaccessible from the internet.

Successful exploitation allows attackers to execute arbitrary commands on the system hosting Ray. Attackers could potentially steal sensitive data, compromise AI/ML workloads, deploy malware, or use the affected infrastructure for cryptocurrency mining. The vulnerability may also allow attackers to target Ray instances located within private corporate networks by using the victim’s browser as an intermediary.

The vulnerability has been addressed in Ray version 2.52.0. CISA’s addition of the vulnerability to its KEV catalog indicates that exploitation has been observed in the wild, increasing the urgency for organizations to identify and remediate affected systems.

Recommendation

Organizations should implement a layered security approach to reduce the risk of Ray exploitation:

• Upgrade Ray to version 2.52.0 or later.
• Avoid exposing Ray management and API endpoints directly to the internet.
• Restrict access to Ray services to trusted hosts and networks using firewalls or network segmentation.
• Monitor Ray servers for unexpected processes, command execution, unusual outbound connections, and cryptocurrency-mining activity.
• Review existing Ray deployments and identify systems running vulnerable versions.

Source

https://thehackernews.com/2026/08/cisa-flags-actively-exploited-ray-flaw.html