CVE-2022-35967: TensorFlow DoS — HIGH

CISO Take

Any TensorFlow deployment exposing inference endpoints that process external tensor inputs is vulnerable to service crashes. An unauthenticated remote attacker can send a crafted request with malformed min/max input tensors to trigger a segfault and take down the inference service. Upgrade to TensorFlow 2.10.0, 2.9.1, 2.8.1, or 2.7.2 immediately — no workaround exists.

What is the risk?

High risk for externally-exposed TensorFlow serving infrastructure. CVSS 7.5 with network-accessible, zero-authentication, low-complexity exploit path means any internet-facing TF endpoint is trivially crashable. Quantized models are commonly deployed in production for performance optimization, widening the exposure surface. No confidentiality or integrity impact, but availability loss of an AI inference service can directly disrupt business operations and SLAs.

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

7.5 / 10

EPSS

0.4%

chance of exploitation in 30 days

Higher than 31% of all CVEs

Source: EPSS v3 — FIRST.org

Exploitation Status

Exploit Available

Exploitation: MEDIUM

Sophistication

Moderate

Exploitation Confidence

medium

○ 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 None

UI None

S Unchanged

C None

I None

A High

What should I do?

5 steps

Patch: Upgrade TensorFlow to 2.10.0, 2.9.1, 2.8.1, or 2.7.2 — commit 49b3824d83af706df0ad07e4e677d88659756d89.
Input validation: Enforce scalar rank (rank=0) on min_input and max_input tensors at API boundaries before passing to QuantizedAdd.
Network controls: Restrict inference API access to trusted networks or authenticated clients only — prevents unauthenticated exploitation.
Detection: Monitor for unexpected service crashes or SIGSEGV signals in TF serving processes; repeated crashes from the same source IP indicate exploitation attempts.
No known workaround beyond patching — treat as P1 if on affected versions.

What does CISA's SSVC say?

Decision Track

Exploitation none

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.T0010.001 - AI Software 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

6.1.2 - AI Risk Assessment 8.4 - AI System Operation

NIST AI RMF

MANAGE 2.2 - Mechanisms are in place to sustain the AI system's trustworthiness

OWASP LLM Top 10

LLM05:2023 - Supply Chain Vulnerabilities

Frequently Asked Questions

What is CVE-2022-35967?

Any TensorFlow deployment exposing inference endpoints that process external tensor inputs is vulnerable to service crashes. An unauthenticated remote attacker can send a crafted request with malformed min/max input tensors to trigger a segfault and take down the inference service. Upgrade to TensorFlow 2.10.0, 2.9.1, 2.8.1, or 2.7.2 immediately — no workaround exists.

Is CVE-2022-35967 actively exploited?

Proof-of-concept exploit code is publicly available for CVE-2022-35967, increasing the risk of exploitation.

How to fix CVE-2022-35967?

1. Patch: Upgrade TensorFlow to 2.10.0, 2.9.1, 2.8.1, or 2.7.2 — commit 49b3824d83af706df0ad07e4e677d88659756d89. 2. Input validation: Enforce scalar rank (rank=0) on min_input and max_input tensors at API boundaries before passing to QuantizedAdd. 3. Network controls: Restrict inference API access to trusted networks or authenticated clients only — prevents unauthenticated exploitation. 4. Detection: Monitor for unexpected service crashes or SIGSEGV signals in TF serving processes; repeated crashes from the same source IP indicate exploitation attempts. 5. No known workaround beyond patching — treat as P1 if on affected versions.

What systems are affected by CVE-2022-35967?

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

What is the CVSS score for CVE-2022-35967?

CVE-2022-35967 has a CVSS v3.1 base score of 7.5 (HIGH). The EPSS exploitation probability is 0.39%.

What is the AI security impact?

Affected AI Architectures

model servinginference pipelinestraining pipelines

MITRE ATLAS Techniques

AML.T0010.001 AI Software

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: 6.1.2, 8.4

NIST AI RMF: MANAGE 2.2

OWASP LLM Top 10: LLM05:2023

What are the technical details?

Original Advisory

TensorFlow is an open source platform for machine learning. If `QuantizedAdd` is given `min_input` or `max_input` tensors of a nonzero rank, it results in a segfault that can be used to trigger a denial of service attack. We have patched the issue in GitHub commit 49b3824d83af706df0ad07e4e677d88659756d89. The fix will be included in TensorFlow 2.10.0. We will also cherrypick this commit on TensorFlow 2.9.1, TensorFlow 2.8.1, and TensorFlow 2.7.2, as these are also affected and still in supported range. There are no known workarounds for this issue.

Exploitation Scenario

An adversary targets a company running quantized TensorFlow models via a public-facing REST inference API (e.g., TensorFlow Serving). They craft a malicious inference request where min_input and max_input tensors are passed as rank-1 or higher tensors instead of scalars. When TensorFlow processes the QuantizedAdd operation, the missing rank validation triggers a segfault, crashing the serving process. The attacker repeats this to sustain a denial of service against the AI inference endpoint, disrupting production ML services without any credentials or prior access.

Weaknesses (CWE)

CWE-20 Improper Input Validation

CWE-20 — Improper Input Validation: The product receives input or data, but it does not validate or incorrectly validates that the input has the properties that are required to process the data safely and correctly.

[Architecture and Design] Consider using language-theoretic security (LangSec) techniques that characterize inputs using a formal language and build "recognizers" for that language. This effectively requires parsing to be a distinct layer that effectively enforces a boundary between raw input and internal data representations, instead of allowing parser code to be scattered throughout the program, where it could be subject to errors or inconsistencies that create weaknesses. [REF-1109] [REF-1110] [REF-1111]
[Architecture and Design] Use an input validation framework such as Struts or the OWASP ESAPI Validation API. Note that using a framework does not automatically address all input validation problems; be mindful of weaknesses that could arise from misusing the framework itself (CWE-1173).

Source: MITRE CWE corpus.