KEV Intelligence is becoming Previdian.

Vulnerability report

Active exploitation observed Confirmed confidence Not in CISA KEV

CVE-2026-2614

mlflow/mlflow Arbitrary File Read Arbitrary File Access

Mlflow / mlflow/mlflow · affected before *

View live feed Monitor via API Watch CVE-2026-2614
Severity
CVSS 7.5 · High
Confidence
Confirmed
Exploit status
Observed in sensors
EPSS
3.0%
First observed
01 Jul 2026
Last observed
29 Aug 2026

Decision summary

What security teams need to know first

Direct answers before the deeper technical record.

What it is

CVE-2026-2614 is an unauthenticated vulnerability affecting Mlflow mlflow/mlflow. A vulnerability in the `_create_model_version()` handler of `mlflow/server/handlers.py` in mlflow/mlflow versions 3.9.0 and earlier...

Is it exploited?

Yes. Previdian sensors observed exploitation attempts with confirmed confidence.

Who is affected?

Mlflow / mlflow/mlflow affected before *.

What should we do?

Patch immediately, validate internet-facing exposure, and monitor for matching requests.

Overview

mlflow/mlflow Arbitrary File Read Arbitrary File Access

A vulnerability in the `_create_model_version()` handler of `mlflow/server/handlers.py` in mlflow/mlflow versions 3.9.0 and earlier allows an unauthenticated remote attacker to read arbitrary files from the server's filesystem. The issue arises when a `CreateModelVersion` request includes the tag `mlflow.prompt.is_prompt`, which bypasses source path validation. This enables an attacker to store an arbitrary local filesystem path as the model version source.

The `get_model_version_artifact_handler()` function later uses this source to serve files without verifying the model version's prompt status, leading to a complete confidentiality compromise. This issue is fixed in version 3.10.0.

Vendor Product Affected Status
Red Hat Red Hat OpenShift AI 3.3 rhoai/odh-training-cuda128-torch29-py312-rhel9 Before * Unaffected
Red Hat Red Hat OpenShift AI 3.4 rhoai/odh-th06-cpu-torch210-py312-rhel9 Before * Unaffected
Red Hat Red Hat OpenShift AI 3.4 rhoai/odh-th06-cuda130-torch210-py312-rhel9 Before * Unaffected
Red Hat Red Hat OpenShift AI 3.4 rhoai/odh-th06-rocm64-torch291-py312-rhel9 Before * Unaffected
Red Hat Red Hat OpenShift AI 3.4 rhoai/odh-training-cuda128-torch29-py312-rhel9 Before * Unaffected
Red Hat Red Hat OpenShift AI 3.4 rhoai/odh-pipeline-runtime-datascience-cpu-py312-rhel9 Before * Unaffected
Red Hat Red Hat OpenShift AI 3.4 rhoai/odh-pipeline-runtime-pytorch-cuda-py312-rhel9 Before * Unaffected
Red Hat Red Hat OpenShift AI 3.4 rhoai/odh-pipeline-runtime-pytorch-llmcompressor-cuda-py312-rhel9 Before * Unaffected
Red Hat Red Hat OpenShift AI 3.4 rhoai/odh-pipeline-runtime-pytorch-rocm-py312-rhel9 Before * Unaffected
Red Hat Red Hat OpenShift AI 3.4 rhoai/odh-pipeline-runtime-tensorflow-cuda-py312-rhel9 Before * Unaffected
Red Hat Red Hat OpenShift AI 3.4 rhoai/odh-pipeline-runtime-tensorflow-rocm-py312-rhel9 Before * Unaffected
Red Hat Red Hat OpenShift AI 3.4 rhoai/odh-workbench-codeserver-datascience-cpu-py312-rhel9 Before * Unaffected
Red Hat Red Hat OpenShift AI 3.4 rhoai/odh-workbench-jupyter-datascience-cpu-py312-rhel9 Before * Unaffected
Red Hat Red Hat OpenShift AI 3.4 rhoai/odh-workbench-jupyter-pytorch-cuda-py312-rhel9 Before * Unaffected
Red Hat Red Hat OpenShift AI 3.4 rhoai/odh-workbench-jupyter-pytorch-rocm-py312-rhel9 Before * Unaffected
Red Hat Red Hat OpenShift AI 3.4 rhoai/odh-workbench-jupyter-tensorflow-cuda-py312-rhel9 Before * Unaffected
Red Hat Red Hat OpenShift AI 3.4 rhoai/odh-workbench-jupyter-tensorflow-rocm-py312-rhel9 Before * Unaffected
Red Hat Red Hat OpenShift AI 3.4 rhoai/odh-workbench-jupyter-trustyai-cpu-py312-rhel9 Before * Unaffected
Red Hat Red Hat OpenShift AI (RHOAI) rhoai/odh-mlflow-rhel9 All versions (default: unaffected) Unaffected
Red Hat Red Hat OpenShift AI (RHOAI) rhoai/odh-workbench-jupyter-pytorch-llmcompressor-cuda-py312-rhel9 All versions (default: affected) Affected
mlflow mlflow/mlflow Before 3.10.0 Affected
Published
11 May 2026
Exploited Since
01 Jul 2026
Attack vector
Remote
Complexity
Low
Privileges
None
User interaction
None

Tags

nuclei_scanner

CVE References

Exploitation evidence

Why Previdian marks this CVE as exploited

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

Exploited in the wild

Previdian

Recorded 20 Aug 2026

Independent exploitation attestation added to the Previdian record.

Active exploitation observed

Previdian sensor

First observed 01 Jul 2026

First-party sensor telemetry confirms matching exploitation attempts.

Proof of concept available

Nuclei Templates

Recorded 15 Aug 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 Pro API.

Learn about Pro API access
Source Added
Previdian First 2026-08-29 20:39 UTC

Operational indicators for this CVE are listed under Detection.

Sensor telemetry

First-party evidence of exploitation activity

Aggregate observations show the scale, recency, and distribution of activity without overstating sparse data.

9

Attempts observed

3

Unique attacker IPs

2

Attacker countries

AF · US

3

Sensors observed

Exploitation attempts over the last 65 days

Daily events observed by Previdian sensors

Updated 02 Sep 2026

0
2
0
52
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
5
0
0
0
5
0
0
0
0
0
0
4
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
5
0
0
0
0
0
0
0
0
4
0
0
0
0
30 Jun 13 Jul 26 Jul 8 Aug 2 Sep

First observed 01 Jul 2026 · Last observed 29 Aug 2026

Pro adds sensor region and window summaries. Enterprise adds raw IPs, paths, User-Agents, and payloads.

Request telemetry access

Detection

Operational artifacts and observed signals

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

Observed signals

Request targets
4
User-Agents
9

Raw values available in Pro and Enterprise.

Virtual patch status

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

Learn about virtual patches →

Attacker IP Indicators

Attacker IP indicators observed · available in Pro and Enterprise.

Sensor-derived attacker IP indicators are available to Pro and Enterprise accounts under Detection and through the Pro API.

Learn about Pro API access

Scanner Artifacts

Scanner and exploit-framework references linked to this CVE.

Scanner Reference Detected
Nuclei https://github.com/projectdiscovery/nuclei-templates/blob/main/http/cves/2026/CVE-2026-2614.yaml 15 Aug 2026

Risk and context

Severity, weaknesses, and research context

CVSS v3.1

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

EPSS

3.0%

Weaknesses

Improper Limitation of a Pathname to a Restricted Directory ('Path Traversal') CWE-22

All CVSS Scores

CVSS v3.1 7.5 High

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

CVSS v3.0 7.5 High

CVSS:3.0/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N

Potential Proof of Concepts

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

MLflow <= 3.9.0 - Arbitrary File Read

nuclei · Created Unknown

Timeline

From disclosure to observed exploitation

  1. Indicators of compromise added (3)

    Indicators of compromise recorded

  2. Public PoC available

    Public proof-of-concept code published

  3. Nuclei template available

    Scanner coverage available

  4. Observed by Previdian sensors

    Evidence-backed exploitation signal

  5. CVE published

    Vulnerability disclosed publicly

  6. CVE ID reserved

    Identifier reserved by the CNA

Pro API

Automate this intelligence

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

View Pro pricing Read API docs
  • Evidence confidence and provenance
  • First-party sensor telemetry
  • PoC and scanner references
  • Affected versions and enrichment
  • Automation-ready JSON delivery

GET /api/v2/pro/kevs/CVE-2026-2614

Free JSON includes basic KEV fields 
{
  "cve_id": "CVE-2026-2614",
  "confidence": "Confirmed",
  "cvss_score": 7.5,
  "cvss_estimated": false,
  "epss_score": 0.02982,
  "exploit_status": {
    "exploited_in_the_wild": true,
    "active_exploitation_observed": true
  },
  "sensor_telemetry": { "attempts": 9, "sensors": 3 }
}