Vulnerability report

Trusted Third Party Report Confirmed Confidence In CISA KEV

CVE-2026-40217

LiteLLM

BerriAI / LiteLLM · affected before *

Severity
CVSS 8.8 · High
Confidence
Confirmed
Exploit status
Trusted Third Party Report
EPSS
17.0%
First observed
—
Last observed
—

Decision summary

What security teams need to know first

Direct answers before the deeper technical record.

What it is

CVE-2026-40217 is a vulnerability affecting BerriAI LiteLLM. LiteLLM through 2026-04-08 allows remote attackers to execute arbitrary code via bytecode rewriting at the /guardrails/test_custom_code URI.

Have exploitation attempts been observed?

Yes. Exploitation attempts have been reported for this vulnerability. Previdian lists it as a known exploited vulnerability. Listed in CISA KEV. Also reported by third-party sources. Confidence is confirmed.

Who is affected?

BerriAI / LiteLLM, including affected before *, affected before *, and All versions (default: affected).

What should we do?

Prioritize remediation. Validate affected assets and apply vendor fixes on an accelerated timeline.

Overview

LiteLLM

LiteLLM through 2026-04-08 allows remote attackers to execute arbitrary code via bytecode rewriting at the /guardrails/test_custom_code URI.

Red Hat Unaffected
Red Hat Ansible Automation Platform 2.6 ansible-automation-platform-26/lightspeed-chatbot-rhel9
Before *
Red Hat Unaffected
Red Hat OpenShift AI 3.3 rhoai/odh-llama-stack-core-rhel9
Before *
Red Hat Affected
All versions (default: affected)
Red Hat Unaffected
Red Hat OpenShift AI (RHOAI) rhoai/odh-mlflow-rhel9
All versions (default: unaffected)
BerriAI Affected
LiteLLM
bb0639701796218a3447160e55c0f1097446e4e6085df7dfd39f476d4143743f
Published
10 Apr 2026
Exploitation Attempts Reported
16 Nov 2026
Attack vector
Remote
Complexity
Low
Privileges
Low
User interaction
None

Tags

ruby nuclei_scanner cisa

CVE References

Exploitation evidence

Evidence of exploitation attempts

Third-party attestation and first-party sensor observation are shown separately so teams can judge the evidence chain.

Exploitation Attempts Observed in the Wild

Google Threat Intelligence

Recorded 16 Nov 2026

A trusted third party reported exploitation attempts.

Proof of concept available

GitHub

Recorded 19 May 2026

Public scanner or PoC coverage increases practical exploitability.

Known exploited vulnerability sources

Per-source evidence links for KEV attestations are available through the Previdian API.

Request a Demo
Source Added
Google Threat Intelligence First 2026-11-16 17:22 UTC
CISA 2026-11-18 19:30 UTC

Detection

Operational artifacts and observed signals

Make the evidence actionable in scanner, SOC, and edge-control workflows.

Virtual patch status

No Previdian virtual patch is currently available. Future rules ship for ModSecurity, Cloudflare, and AWS WAF.

Learn about virtual patches →

Scanner Artifacts

Scanner and exploit-framework references linked to this CVE.

Risk and context

Severity, weaknesses, and research context

CVSS v3.1

8.8 High

Potential damage if exploited. Separate from whether attackers are using it.

CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H

EPSS

17.0%

Estimated chance of exploitation in the next 30 days. Previdian's warning comes from evidence, not this score.

Recent mention · Security Affairs

Google GTIG Warns AI Agents Automate Exploits, Steal Credentials and Scale Cybercrime

Google finds attackers using autonomous AI agents to automate exploits, steal AI access and scale intrusions, often sacrificing stealth for speed. Google’s Threat Intelligence Group (GTIG) says attackers have moved past typing prompts into a chatbot. They now run autonomous,...

Read full advisory

All Mentions

Recent mention · Security Affairs

Google GTIG Warns AI Agents Automate Exploits, Steal Credentials and Scale Cybercrime

Security Affairs · 16 Nov 2026

Google finds attackers using autonomous AI agents to automate exploits, steal AI access and scale intrusions, often sacrificing stealth for speed. Google’s Threat Intelligence Group (GTIG) says attackers have moved past typing prompts into a chatbot. They now run autonomous, multi-agent frameworks such as ARTEX, CAIRN, CyberStrikeAI and PentAGI to carry out multi-stage intrusions faster. […]

Recent mention · Google Threat Intelligence

Written by: Aurora Blum, Kelli Vanderlee, Jess Xia, David Mainor Executive Summary  As threat actors increasingly move beyond basic prompting to...

Google Threat Intelligence · 16 Nov 2026

Written by: Aurora Blum, Kelli Vanderlee, Jess Xia, David Mainor Executive Summary  As threat actors increasingly move beyond basic prompting to deploying autonomous multi-agent frameworks, automated vulnerability exploitation pipelines, and alternative monetization schemes like LLMJacking, the adversarial AI landscape has reached a critical inflection point. In this update, Google Threat Intelligence Group (GTIG) and Mandiant examine how adversaries across the cyber crime, espionage, and information operations spectrum are operationalizing generative AI to compress attack timelines, scale...

Recent mention · The Hacker News

Google Warns LLMjacking Attacks Are Costing Some Enterprises Up to $1 Million in Cloud Charges

The Hacker News · 03 Nov 2026

Ravie LakshmananNov 03, 2026Cloud Security / Artificial Intelligence Google on Monday warned that threat actors are hijacking enterprise access to Large Language Models (LLMs) to fraudulently leverage expensive computing resources and AI services. The tactic is called LLMjacking, a form of resource hijacking adapted for the LLM era, where an attacker obtains an API key through phishing, source code leaks, misconfigured environments, or exposed client-side code to gain unauthorized access to an organization's LLM accounts, while the legitimate owner covers the consumption costs. Such...

Recent mention · Google Threat Intelligence

Written by: Christopher Romano, Codi Starks, Garrett Mahoney, John Wolfram, Josh Murchie, K Singh, Mathew Potaczek, Mike Lombardi, Rik...

Google Threat Intelligence · 02 Nov 2026

Written by: Christopher Romano, Codi Starks, Garrett Mahoney, John Wolfram, Josh Murchie, K Singh, Mathew Potaczek, Mike Lombardi, Rik Dolfing Introduction  Mandiant responded to multiple incidents in 2026 in which threat actors hijacked enterprise access to Large Language Models (LLMs), a tactic known as "LLMJacking." Spanning a variety of cloud and LLM providers, these intrusions impacted organizations in North America, South America, and Asia Pacific across the financial services, retail, technology, and transportation sectors.  Unlike the gradual resource consumption associated with...

Potential Proof of Concepts

These PoCs are unverified and could contain malware. Use at your own risk.

Timeline

From disclosure to observed exploitation attempts

  1. 19:30 UTC

    Added to CISA KEV

    Listed in the CISA Known Exploited Vulnerabilities catalog

  2. 17:22 UTC

    Added to Previdian KEV Feed

    High-confidence, third-party report of exploitation attempts

  3. 04:31 UTC

    Public PoC available

    Public proof-of-concept code published

  4. 04:31 UTC

    Nuclei template available

    Scanner coverage available

  5. 13:43 UTC

    CVE published

    Vulnerability disclosed publicly

  6. 13:43 UTC

    CVE ID reserved

    Identifier reserved by the CNA

API

Automate this intelligence

Confidence, exploit status, sensor telemetry, PoCs, scanner integrations, mentions, and tags are available programmatically for VM, SOC, and CTI workflows.

  • Evidence confidence and provenance
  • First-party sensor telemetry
  • PoC and scanner references
  • Affected versions and enrichment
  • Automation-ready JSON delivery

GET /api/v2/enterprise/kevs/CVE-2026-40217

Sample API response
{
  "cve_id": "CVE-2026-40217",
  "confidence": "Confirmed",
  "cvss_score": 8.8,
  "cvss_estimated": false,
  "epss_score": 0.1702,
  "exploit_status": {
    "exploited_in_the_wild": true,
    "active_exploitation_observed": false
  },
  "sensor_telemetry": { "attempts": 0, "sensors": 0 }
}