CVE-2021-29562: TensorFlow assertion failure DoS

CISO Take

A local attacker with low privileges can crash TensorFlow processes by triggering a CHECK assertion failure in tf.raw_ops.IRFFT with malformed inputs. Risk is concentrated in shared ML platforms (JupyterHub, Kubeflow) where multiple users share compute resources. Patch to TF 2.5.0, 2.4.2, 2.3.3, 2.2.3, or 2.1.4 during next maintenance window — not an emergency but should not remain unpatched.

What is the risk?

CVSS 5.5 (Medium) with local attack vector limits internet-facing exposure significantly. In shared ML infrastructure — common in enterprise data science teams — this is exploitable by any authenticated local user with minimal TF knowledge. Impact is availability-only (A:H) with zero confidentiality or integrity risk. Priority is moderate: schedule patching but not emergency response unless operating a multi-tenant ML platform with untrusted users.

What systems are affected?

Package	Ecosystem	Vulnerable Range	Patched
TensorFlow	pip	—	No patch
195.8K OpenSSF 7.1 3.7K dependents Pushed 5d ago 4% patched ~1372d to patch Full package profile →

Do you use TensorFlow? You're affected.

How severe is it?

CVSS 3.1

5.5 / 10

EPSS

0.2%

chance of exploitation in 30 days

Higher than 9% of all CVEs

Source: EPSS v3 — FIRST.org

Exploitation Status

Exploit Available

Exploitation: MEDIUM

Sophistication

Trivial

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 Local

AC Low

PR Low

UI None

S Unchanged

C None

I None

A High

What should I do?

4 steps

Patch: Upgrade to TensorFlow 2.5.0, 2.4.2, 2.3.3, 2.2.3, or 2.1.4. Apply commit 1c56f53be0b722ca657cbc7df461ed676c8642a2 as a hotfix if immediate upgrade is not feasible.
Validate inputs: Add dtype and shape validation before any IRFFT operation call in production code.
Restrict graph submission: In shared environments, limit which users can submit custom computation graphs.
Detect: Monitor for TF process crashes and CHECK assertion failure log entries; alert on repeated failures from the same user or session.

How is it classified?

DoS Framework AML.T0010.001 - AI Software AML.T0029 - Denial of AI Service

Which compliance frameworks are affected?

This CVE is relevant to:

EU AI Act

Article 9 - Risk management system

ISO 42001

A.6.2.4 - AI System Operation

NIST AI RMF

MANAGE-2.2 - Mechanisms to sustain the value of deployed AI systems

OWASP LLM Top 10

LLM04 - Model Denial of Service

Frequently Asked Questions

What is CVE-2021-29562?

A local attacker with low privileges can crash TensorFlow processes by triggering a CHECK assertion failure in tf.raw_ops.IRFFT with malformed inputs. Risk is concentrated in shared ML platforms (JupyterHub, Kubeflow) where multiple users share compute resources. Patch to TF 2.5.0, 2.4.2, 2.3.3, 2.2.3, or 2.1.4 during next maintenance window — not an emergency but should not remain unpatched.

Is CVE-2021-29562 actively exploited?

Proof-of-concept exploit code is publicly available for CVE-2021-29562, increasing the risk of exploitation.

How to fix CVE-2021-29562?

1. Patch: Upgrade to TensorFlow 2.5.0, 2.4.2, 2.3.3, 2.2.3, or 2.1.4. Apply commit 1c56f53be0b722ca657cbc7df461ed676c8642a2 as a hotfix if immediate upgrade is not feasible. 2. Validate inputs: Add dtype and shape validation before any IRFFT operation call in production code. 3. Restrict graph submission: In shared environments, limit which users can submit custom computation graphs. 4. Detect: Monitor for TF process crashes and CHECK assertion failure log entries; alert on repeated failures from the same user or session.

What systems are affected by CVE-2021-29562?

This vulnerability affects the following AI/ML architecture patterns: training pipelines, model serving, audio/signal processing ML pipelines, shared notebook environments.

What is the CVSS score for CVE-2021-29562?

CVE-2021-29562 has a CVSS v3.1 base score of 5.5 (MEDIUM). The EPSS exploitation probability is 0.19%.

What is the AI security impact?

Affected AI Architectures

training pipelinesmodel servingaudio/signal processing ML pipelinesshared notebook environments

MITRE ATLAS Techniques

AML.T0010.001 AI Software

AML.T0029 Denial of AI Service

Compliance Controls Affected

EU AI Act: Article 9

ISO 42001: A.6.2.4

NIST AI RMF: MANAGE-2.2

OWASP LLM Top 10: LLM04

What are the technical details?

Original Advisory

TensorFlow is an end-to-end open source platform for machine learning. An attacker can cause a denial of service by exploiting a `CHECK`-failure coming from the implementation of `tf.raw_ops.IRFFT`. The fix will be included in TensorFlow 2.5.0. We will also cherrypick this commit on TensorFlow 2.4.2, TensorFlow 2.3.3, TensorFlow 2.2.3 and TensorFlow 2.1.4, as these are also affected and still in supported range.

Exploitation Scenario

An attacker with local access to a shared JupyterHub or Kubeflow environment writes a Python snippet calling tf.raw_ops.IRFFT with an invalid tensor — wrong dtype or mismatched shape triggers the internal CHECK assertion. This causes an unhandled exception crashing the TensorFlow runtime process. On a shared notebook server, this interrupts other users' long-running training jobs. The attack requires only basic TensorFlow knowledge: identifying the vulnerable op and supplying a malformed input.

Weaknesses (CWE)

CWE-617 Reachable Assertion

CWE-617 — Reachable Assertion: The product contains an assert() or similar statement that can be triggered by an attacker, which leads to an application exit or other behavior that is more severe than necessary.

[Implementation] Make sensitive open/close operation non reachable by directly user-controlled data (e.g. open/close resources)
[Implementation] Perform input validation on user data.

Source: MITRE CWE corpus.