GCD AI RADAR
incidents/AIR-2025-0049

ShadowRay Active Exploitation of Exposed Ray Clusters

Attack on AI3 Confirmed exploitationInfrastructure vulnerabilityAgent hijack

Threat actors actively exploited unauthenticated and exposed Ray clusters over several months for remote code execution. The attacks compromised significant computing resources and valuable enterprise data.

Details

type
Attack on AI
An AI model, agent or AI-powered product was the thing exploited, abused or leaked.
categories
  • Infrastructure vulnerabilityprimary
    CVEs and advisories in AI frameworks, inference servers, MCP servers, AI developer tools and hosting stacks.
  • Agent hijack
    An AI agent is tricked into unauthorized actions: tool misuse, data exfiltration through the agent, or acting against its principal.
owasp llm top 10
mitre atlas
Both rows are derived from the type and categories (OWASP 2025 edition, ATLAS release 2026.09), not classified per record. How
severity
3 · Confirmed exploitation
Used against real targets; limited or unclear impact.
targets
Framework / tooling
products
Ray
incident date
5 Sep 2023
first reported
28 Nov 2023
origin
MITRE ATLAS
Derived from MITRE ATLAS (Apache 2.0, © MITRE).
classifier confidence
0.95 · published at or above 0.80; see the methodology
tags
shadowray · ray · remote-code-execution · CVE-2023-48022
record
created 2026-10-18 06:55Z · updated 2026-10-18 07:00Z · json

Sources · 6

  1. 2023-11-28
    CVE-2023-48022: Ray has arbitrary code execution via jobs submission API
    GitHub Advisory Database · github.com/advisories/GHSA-6wgj-66m2-xxp2
  2. 2025-03-14
    ShadowRay: Hijacking Exposed Ray Clusters
    MITRE ATLAS · atlas.mitre.org/studies/AML.CS0023
  3. 2025-03-14
    Anyscale Update on CVEs
    anyscale.com · anyscale.com/blog/update-on-ray-cves-cve-2023-6019-cve-2023-6020-cve-2023-6021-c
  4. 2025-03-14
    CVE-2023-48022
    NVD · nvd.nist.gov/vuln/detail/CVE-2023-48022
  5. 2025-03-14
    ShadowRay: First Known Attack Campaign Targeting AI Workloads Actively Exploited In The Wild
    oligo.security · oligo.security/blog/shadowray-attack-ai-workloads-actively-exploited-in-the-wild
  6. 2025-03-14
    ShadowRay: AI Infrastructure Is Being Exploited In the Wild
    Protect AI · protectai.com/threat-research/shadowray-ai-infrastructure-is-being-exploited-in-

Titles and links belong to their publishers. The summary above is written by GCD AI RADAR in its own words; data is CC BY-SA 4.0.

Related incidents