LiveActive security incident?Get immediate response
CVE Record

CVE-2021-29608: Heap OOB and null pointer dereference in `RaggedTensorToTensor`

TensorFlow is an end-to-end open source platform for machine learning. Due to lack of validation in `tf.raw_ops.RaggedTensorToTensor`, an attacker can exploit an undefined behavior if input arguments are empty. The implementation(https://github.com/tensorflow/tensorflow/blob/656e7673b14acd7835dc778867f84916c6d1cac2/tensorflow/core/kernels/ragged_tensor_to_tensor_op.cc#L356-L360) only checks that one of the tensors is not empty, but does not check for the other ones. There are multiple `DCHECK` validations to prevent heap OOB, but these are no-op in release builds, hence they don't prevent anything. The fix will be included in TensorFlow 2.5.0. We will also cherrypick these commits 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.

MediumCVSS 5.3Not KEV-listedUpdated
Discuss exposureBack to CVEs
Glexia's TakeAutomated analysismoderate

Security readout for executives and security teams

Plain-English summary

This TensorFlow flaw can make vulnerable ML workloads crash or behave incorrectly when empty RaggedTensor conversion inputs are processed. The main business risk is service disruption in systems that process user-controlled tensor data. The bundle does not show known active exploitation or identify affected downstream products beyond TensorFlow.

Executive priority

Prioritize patching in shared or user-facing ML environments because the strongest documented impact is availability loss. For isolated research systems with trusted inputs, schedule normal dependency remediation. Do not treat this as an emergency based on the supplied evidence.

Technical view

A validation gap in `tf.raw_ops.RaggedTensorToTensor` can trigger heap out-of-bounds behavior and null pointer dereference when input arguments are empty. Debug-only `DCHECK` validations did not protect release builds. CVSS 3.1 is 5.3: local attack vector, high complexity, low privileges, low integrity impact, and high availability impact.

Likely exposure

Affected TensorFlow versions are before 2.1.4, 2.2.0 to before 2.2.3, 2.3.0 to before 2.3.3, and 2.4.0 to before 2.4.2. Exposure is most relevant where users or tenants can influence tensor inputs in ML pipelines, notebooks, batch jobs, or APIs.

Exploitation context

The CVE is not in KEV, and the provided sources do not claim active exploitation. The CVSS vector indicates local access, low privileges, no user interaction, and high attack complexity. Treat exploitation likelihood as lower than network-reachable flaws, but availability impact can matter for shared ML services.

Researcher notes

The root issue is incomplete empty-input validation in `RaggedTensorToTensor`; release builds were not protected by debug-only assertions. Evidence supports affected TensorFlow versions and fixed release targets, but does not provide downstream product impact, active exploitation, or a standalone workaround beyond upgrading.

Mitigation direction

  • Upgrade TensorFlow to 2.5.0 or a patched supported branch release.
  • Use TensorFlow 2.4.2, 2.3.3, 2.2.3, or 2.1.4 where pinned to those branches.
  • Inventory ML runtimes, containers, notebooks, and training images for vulnerable TensorFlow versions.
  • Restrict untrusted access to workflows that accept arbitrary tensor inputs until patched.
  • Monitor vendor guidance for any downstream package or platform-specific remediation.

Validation and detection

  • Check dependency lockfiles and runtime environments for TensorFlow version ranges listed as affected.
  • Review SBOMs and container images for vulnerable TensorFlow packages.
  • Identify code paths using `tf.raw_ops.RaggedTensorToTensor` or processing user-controlled ragged tensors.
  • Confirm patched versions are deployed in production, CI, notebooks, and batch workers.
  • Run existing ML regression tests after upgrade to catch compatibility issues.
Prepared
Confidence
high
Sources
6

Public sources used

Generated from the cited source records. This long-tail analysis has not been individually reviewed by a named human.

Potential ATT&CK relevance

Conservative CVE-to-ATT&CK context

These mappings and lookup hints may be relevant to the vulnerability behavior, CWE, affected product, or exposure path. Glexia-inferred context is not an official MITRE, ATT&CK, CWE, or CVE Program mapping.

ATT&CK lookup starting points

Use these exact CWE pages and searches to review the Glexia ATT&CK library from this CVE's weakness and description context.

cwe · low confidence lookup

CWE-131: Exact CWE lookup

Use the exact CWE identifier as the starting point before reviewing related ATT&CK behavior. Open the exact CWE lookup page first, then review the ATT&CK searches from that MITRE weakness context. This is a Glexia lookup hint, not an official ATT&CK mapping.

Open ATT&CK lookup
cve · low confidence lookup

CVE-2021-29608 mapping review

Open the CVE-to-ATT&CK bridge for reviewed, inferred, or future official mappings tied to this CVE.

Open ATT&CK lookup
Vulnerability profileCVE Program record
Severity
Medium
CVSS
5.3 (3.1)
Known Exploited
No
Published

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

Official CVE source material

CNA and ADP enrichment extracted from CVE v5

These fields come from the CVE record and ADP containers, not from Glexia's Take. They preserve time-varying source decisions such as CISA SSVC, KEV status, CVSS metrics, and provider references.

1CVSS vectors
0Timeline events
0ADP providers
5Source links

CVSS vector scores

1 official score

We collect every scored CVSS vector available in the official CNA and ADP containers. When more than one version is present, the table keeps the source vectors side by side instead of collapsing them into the highest score.

ScoreVersionSeverityVectorExploitImpactSource
5.3CVSS 3.1MediumCVSS:3.1/AV:L/AC:H/PR:L/UI:N/S:U/C:N/I:L/A:H14.2Primary CVE score

Vulnerability scoring details

Base CVSS 3.1 score

5.3Medium
CVSS 3.1 vector shape for CVE-2021-29608Attack VectorAttack ComplexityPrivileges RequiredUser InteractionScopeConfidentiality ImpactIntegrity ImpactAvailability Impact

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

Attack Vector
NetworkAdjacentLocalPhysical
Attack Complexity
LowHigh
Privileges Required
NoneLowHigh
User Interaction
NoneRequired
Scope
ChangedUnchanged
Confidentiality Impact
HighLowNone
Integrity Impact
HighLowNone
Availability Impact
HighLowNone

Source materials

Affected products

Products and packages named in the record

VendorProductVersion / packageStatus
tensorflowtensorflow< 2.1.4, >= 2.2.0, < 2.2.3, >= 2.3.0, < 2.3.3, >= 2.4.0, < 2.4.2Listed
Weakness

CWE details

CWE-131

CWE links open Glexia weakness intelligence pages with official CWE context, developer remediation guidance, and related CVE mappings.

CWE-131 · source CWE mapping

Incorrect Calculation of Buffer Size

Incorrect Calculation of Buffer Size represents a recurring weakness pattern that can create exploitable paths when design, validation, or implementation controls are missing.

Open CWE intelligenceSame-CWE CVEs