CVE-2026-72649

Awaiting Analysis Awaiting Analysis - Queue

Deserialization of Untrusted Data in Elasticsearch ML Leading to RCE

Publication date: 2026-09-01

Last updated on: 2026-09-01

Assigner: Elastic

Description

Deserialization of Untrusted Data (CWE-502) in the Elasticsearch machine learning component can lead to remote code execution via Object Injection (CAPEC-586). A specially crafted trained model artifact could cause attacker-controlled logic to execute with a materially broader system-call surface than intended. Exploitation requires an authenticated user with sufficient privileges to create and deploy trained models.

CVSS Scores

EPSS Scores

Probability:
Percentile:

Meta Information

Published

2026-09-01

Last Modified

2026-09-01

Generated

2026-09-02

AI Q&A

2026-09-02

EPSS Evaluated

N/A

NVD

CVE-2026-72649

EUVD

EUVD-2026-69562

Affected Vendors & Products

Vendor	Product	Version / Range
elastic	elasticsearch	to 8.19.20 (exc)
elastic	elasticsearch	to 9.4.5 (exc)
elastic	elasticsearch	to 9.5.1 (exc)
elastic	elasticsearch	*

Helpful Resources

Exploitability

CWE

KEV

CWE ID	Description
CWE-502	The product deserializes untrusted data without sufficiently ensuring that the resulting data will be valid.

Attack-Flow Graph

Executive Summary

This vulnerability is a deserialization flaw in Elasticsearch's machine learning component that allows remote code execution through object injection. It occurs when an attacker uploads a specially crafted trained model artifact, exploiting CWE-502 to execute attacker-controlled logic with elevated system access. Exploitation requires an authenticated user with sufficient privileges to create and deploy models.

Detection Guidance

Check for unrecognized or untrusted trained models in your Elasticsearch cluster. Review logs for unusual model uploads or deployments. Use Elasticsearch APIs to list deployed models and verify their sources.

Impact Analysis

If exploited, this vulnerability could allow an attacker to execute arbitrary code on the affected Elasticsearch server with high privileges, potentially leading to data breaches, system compromise, or unauthorized access to sensitive information. Clusters without machine learning enabled are not at risk.

Compliance Impact

This vulnerability could lead to unauthorized data access or exfiltration, violating compliance requirements under GDPR (data protection) and HIPAA (health information security). Organizations must address it to maintain regulatory compliance and avoid potential fines or legal consequences.

Mitigation Strategies

Upgrade Elasticsearch to versions 8.19.20, 9.4.5, or 9.5.1. If upgrading is not possible, disable the machine learning feature cluster-wide or avoid uploading third-party trained models.

Hi! I’m here to help you understand CVE-2026-72649. Ask me anything about the vulnerability, its impact, or mitigation strategies.

0/70