CVE-2022-23576: TensorFlow integer overflow

CISO Take

A network-accessible integer overflow in TensorFlow's Grappler cost estimator allows low-privileged attackers to crash TensorFlow workloads via crafted tensor operations. Patch to TF 2.8.0, 2.7.1, 2.6.3, or 2.5.3 immediately if running older versions. Restrict API access to TensorFlow Serving endpoints to minimize exposure surface.

What is the risk?

Medium risk in most environments. CVSS 6.5 reflects network accessibility with low privilege requirements, but impact is limited to availability—no data exfiltration or code execution is possible. Risk increases substantially for multi-tenant AI platforms or shared inference infrastructure where untrusted users can submit custom graph operations. Not in CISA KEV and no known active exploitation as of publication.

What systems are affected?

Package	Ecosystem	Vulnerable Range	Patched
TensorFlow	pip	—	No patch
195.8K OpenSSF 7.1 3.7K dependents Pushed 2d ago 4% patched ~1372d to patch Full package profile →

Do you use TensorFlow? You're affected.

How severe is it?

CVSS 3.1

6.5 / 10

EPSS

0.8%

chance of exploitation in 30 days

Higher than 51% of all CVEs

Source: EPSS v3 — FIRST.org

Exploitation Status

Exploit Available

Exploitation: MEDIUM

Sophistication

Moderate

Exploitation Confidence

medium

○ CISA SSVC: Public PoC

○ Public PoC indexed (trickest/cve)

Composite signal derived from CISA KEV, VulnCheck KEV, CISA SSVC, EPSS, Metasploit, Exploit-DB, trickest/cve, Nuclei templates, and inthewild.io exploitation reports.

What is the attack surface?

AV Network

AC Low

PR Low

UI None

S Unchanged

C None

I None

A High

What should I do?

5 steps

Upgrade to TensorFlow 2.8.0, 2.7.1, 2.6.3, or 2.5.3—all contain the upstream patch (commit b9bd6cfd).
Enforce strict authentication and authorization on all TF Serving and training API endpoints—no unauthenticated tensor submission.
Implement input validation at the API gateway layer to reject operations with pathologically large tensor dimensions.
Set resource quotas on tensor dimension sizes in multi-tenant environments.
Monitor for abnormally large graph submissions or repeated crash/restart cycles as detection signals for exploitation attempts.

What does CISA's SSVC say?

Decision Track*

Exploitation poc

Automatable No

Technical Impact partial

Source: CISA Vulnrichment (SSVC v2.0). Decision based on the CISA Coordinator decision tree.

How is it classified?

DoS Framework Inference AML.T0029 - Denial of AI Service AML.T0043 - Craft Adversarial Data AML.T0049 - Exploit Public-Facing Application

Which compliance frameworks are affected?

This CVE is relevant to:

EU AI Act

Article 15 - Accuracy, robustness and cybersecurity

ISO 42001

A.6.2.6 - AI system availability and resilience

NIST AI RMF

MANAGE-2.2 - Mechanisms to sustain AI system operation and performance

Frequently Asked Questions

What is CVE-2022-23576?

A network-accessible integer overflow in TensorFlow's Grappler cost estimator allows low-privileged attackers to crash TensorFlow workloads via crafted tensor operations. Patch to TF 2.8.0, 2.7.1, 2.6.3, or 2.5.3 immediately if running older versions. Restrict API access to TensorFlow Serving endpoints to minimize exposure surface.

Is CVE-2022-23576 actively exploited?

Proof-of-concept exploit code is publicly available for CVE-2022-23576, increasing the risk of exploitation.

How to fix CVE-2022-23576?

1. Upgrade to TensorFlow 2.8.0, 2.7.1, 2.6.3, or 2.5.3—all contain the upstream patch (commit b9bd6cfd). 2. Enforce strict authentication and authorization on all TF Serving and training API endpoints—no unauthenticated tensor submission. 3. Implement input validation at the API gateway layer to reject operations with pathologically large tensor dimensions. 4. Set resource quotas on tensor dimension sizes in multi-tenant environments. 5. Monitor for abnormally large graph submissions or repeated crash/restart cycles as detection signals for exploitation attempts.

What systems are affected by CVE-2022-23576?

This vulnerability affects the following AI/ML architecture patterns: training pipelines, model serving, MLOps platforms.

What is the CVSS score for CVE-2022-23576?

CVE-2022-23576 has a CVSS v3.1 base score of 6.5 (MEDIUM). The EPSS exploitation probability is 0.77%.

What is the AI security impact?

Affected AI Architectures

training pipelinesmodel servingMLOps platforms

MITRE ATLAS Techniques

AML.T0029 Denial of AI Service

AML.T0043 Craft Adversarial Data

AML.T0049 Exploit Public-Facing Application

Compliance Controls Affected

EU AI Act: Article 15

ISO 42001: A.6.2.6

NIST AI RMF: MANAGE-2.2

What are the technical details?

Original Advisory

Tensorflow is an Open Source Machine Learning Framework. The implementation of `OpLevelCostEstimator::CalculateOutputSize` is vulnerable to an integer overflow if an attacker can create an operation which would involve tensors with large enough number of elements. We can have a large enough number of dimensions in `output_shape.dim()` or just a small number of dimensions being large enough to cause an overflow in the multiplication. The fix will be included in TensorFlow 2.8.0. We will also cherrypick this commit on TensorFlow 2.7.1, TensorFlow 2.6.3, and TensorFlow 2.5.3, as these are also affected and still in supported range.

Exploitation Scenario

An attacker with low-privilege API credentials to a TensorFlow Serving endpoint or shared MLOps platform constructs a computation graph containing operations whose output tensors have very large numbers of dimensions or dimension sizes. When TensorFlow's Grappler optimizer invokes CalculateOutputSize to estimate operation costs, the multiplication of dimension values overflows a 32-bit integer, causing an abort or crash of the TF runtime. In a shared ML training cluster, this is repeatable on demand: the attacker can continuously restart crashed workers, disrupting legitimate training jobs and degrading inference SLAs for all tenants on the platform.

Weaknesses (CWE)

CWE-190 Integer Overflow or Wraparound

CWE-190 — Integer Overflow or Wraparound: The product performs a calculation that can produce an integer overflow or wraparound when the logic assumes that the resulting value will always be larger than the original value. This occurs when an integer value is incremented to a value that is too large to store in the associated representation. When this occurs, the value may become a very small or negative number.

[Requirements] Ensure that all protocols are strictly defined, such that all out-of-bounds behavior can be identified simply, and require strict conformance to the protocol.
[Requirements] Use a language that does not allow this weakness to occur or provides constructs that make this weakness easier to avoid. If possible, choose a language or compiler that performs automatic bounds checking.

Source: MITRE CWE corpus.