Cyber Resilience

CVE-2024-6983

HighPublic PoCRCE

Published: 27 September 2024

Published
27 September 2024
Modified
10 July 2025
KEV Added
Patch
CVSS Score v3 8.8 CVSS:3.0/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
EPSS Score 0.0495 89.9th percentile
Risk Priority 21 60% EPSS · 20% KEV · 20% CVSS

Summary

CVE-2024-6983 is a high-severity Code Injection (CWE-94) vulnerability in Mudler Localai. Its CVSS base score is 8.8 (High).

Operationally, exploitation aligns with the MITRE ATT&CK technique Exploit Public-Facing Application (T1190); ranked in the top 10.1% of CVEs by exploit likelihood; it is not currently listed in the CISA KEV catalog; a public proof-of-concept is referenced.

This vulnerability is AI-related — categorised as APIs and Models; in the Supply Chain and Deployment risk domain; MITRE ATLAS techniques in scope: AML.T0040.000, Invert AI Model (AML.T0024.001), Financial Harm (AML.T0048.000).

Deeper analysis

mudler/localai version 2.17.1 contains a remote code execution vulnerability in its backend component. The flaw stems from the backend accepting inputs from sources other than the configuration file, which enables an attacker to supply a malicious binary. This is tracked as CWE-94 and carries a CVSS 3.0 score of 8.8.

An attacker with limited privileges can send crafted inputs over the network to upload and execute arbitrary code, resulting in full control of the host system. No user interaction is required once the attacker has network access to the LocalAI instance.

The referenced commit d02a0f6f01d5c4a926a2d67190cb55d7aca23b66 addresses the issue, and details are available in the associated huntr.com bounty report. The EPSS score reached a peak of 0.0648 before receding to the current value of 0.0495.

EU & UK References

Vulnerability details

mudler/localai version 2.17.1 is vulnerable to remote code execution. The vulnerability arises because the localai backend receives inputs not only from the configuration file but also from other inputs, allowing an attacker to upload a binary file and execute malicious…

more

code. This can lead to the attacker gaining full control over the system.

CWE(s)

AI Security AnalysisAI

AI Category
APIs and Models
Risk Domain
Supply Chain and Deployment
OWASP Top 10 for LLMs 2025
None mapped
Classification Reason
LocalAI is an open-source, self-hosted API compatible with OpenAI for local inference of AI models including LLMs, making it an APIs and Models platform. The RCE vulnerability in its backend input handling confirms AI-related impact.

Related Threats

MITRE ATT&CK Enterprise TechniquesAI

T1190 Exploit Public-Facing Application Initial Access
Adversaries may attempt to exploit a weakness in an Internet-facing host or system to initially access a network.
Why these techniques?

The vulnerability in mudler/localai enables remote code execution by allowing attackers to upload and execute arbitrary binary files via backend inputs on the public-facing API server.

MITRE ATLAS TechniquesAI

MITRE ATLAS techniques

AML.T0040.000AML.T0024.001: Invert AI ModelAML.T0048.000: Financial Harm

Affected Assets

mudler
localai
2.17.1

Mitigating Controls

Likely Mitigating Controls AI

Per-CVE control mapping for this CVE has not run yet; the list below is derived from the weakness types (CWEs) cited in the NVD entry.

addresses: CWE-94

Makes persistent code injection into loaded programs impossible when the executable image itself resides on hardware-protected read-only media.

addresses: CWE-94

Dynamically generated code can be produced and executed inside the isolated chamber, preventing host compromise from code-injection payloads.

addresses: CWE-94

Validates inputs used in dynamic code generation to block injected directives.

addresses: CWE-94

Directly prevents execution of attacker-supplied code written into data memory regions.

References