Security readout for executives and security teams
Plain-English summary
TensorFlow versions in the affected ranges can behave unsafely when a specific sparse-tensor operation receives empty tensor inputs. In systems where a lower-privileged user can run TensorFlow workloads or influence tensors reaching that operation, this could affect confidentiality, integrity, or availability. Sources do not show active exploitation.
Executive priority
Prioritize patching where TensorFlow is used in shared ML infrastructure or user-facing data-processing services. The issue is high severity, but the supplied evidence points to local or workload-level access rather than unauthenticated remote exploitation.
Technical view
CVE-2021-37676 is a TensorFlow undefined-behavior flaw in shape inference for `tf.raw_ops.SparseFillEmptyRows`. The implementation failed to validate non-empty input tensors, allowing a reference to bind to a null pointer. The advisory rates it CVSS 7.8 with local attack vector and low privileges required.
Likely exposure
Exposure is most likely in ML platforms, notebooks, batch jobs, or services running affected TensorFlow releases where untrusted or lower-privileged users can execute TensorFlow operations or influence tensor inputs. The listed affected ranges are TensorFlow 2.5.0 before 2.5.1, 2.4.x before 2.4.3, and versions before 2.3.4.
Exploitation context
The public sources describe local, low-privilege exploitation potential and no required user interaction. They do not provide evidence of internet-scale exploitation, known malware use, or CISA KEV listing. Treat exploitation status as unconfirmed from the supplied evidence.
Researcher notes
The core issue is missing empty-tensor validation in TensorFlow sparse operation shape inference. The vendor identified the fix commit and target fixed releases. The sources do not name a workaround beyond upgrading, and they do not establish exploitation in the wild.
Mitigation direction
- Upgrade TensorFlow to 2.6.0 or a fixed supported release.
- For 2.5.x, update to TensorFlow 2.5.1 or later.
- For 2.4.x, update to TensorFlow 2.4.3 or later.
- For 2.3.x or older, update to TensorFlow 2.3.4 or a newer fixed release.
- Restrict untrusted users from running arbitrary TensorFlow workloads until patched.
Validation and detection
- Inventory TensorFlow versions in applications, notebooks, containers, and ML images.
- Confirm deployed versions are outside the affected ranges listed by the advisory.
- Review code paths using `tf.raw_ops.SparseFillEmptyRows` with untrusted inputs.
- Verify patched builds include commit `578e634b4f1c1c684d4b4294f9e5281b2133b3ed`.
- Record no active exploitation evidence unless new trusted sources show otherwise.
Public sources used
Generated from the cited source records. This long-tail analysis has not been individually reviewed by a named human.
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-824: 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 lookupCVE-2021-37676 mapping review
Open the CVE-to-ATT&CK bridge for reviewed, inferred, or future official mappings tied to this CVE.
Open ATT&CK lookup- Severity
- High
- CVSS
- 7.8 (3.1)
- Known Exploited
- No
- Published
Vector: CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
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.
CVSS vector scores
1 official scoreWe 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.
CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H1.85.9Primary CVE scoreVulnerability scoring details
Base CVSS 3.1 score
7.8HighVector: CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
Source materials
- CVE List V5 sourceCVE List V5
- https://github.com/tensorflow/tensorflow/security/advisories/GHSA-v768-w7m9-2vmmCVE reference · x_refsource_CONFIRM
- https://github.com/tensorflow/tensorflow/commit/578e634b4f1c1c684d4b4294f9e5281b2133b3edCVE reference · x_refsource_MISC
Products and packages named in the record
CWE details
CWE links open Glexia weakness intelligence pages with official CWE context, developer remediation guidance, and related CVE mappings.
Access of Uninitialized Pointer
Access of Uninitialized Pointer represents a recurring weakness pattern that can create exploitable paths when design, validation, or implementation controls are missing.