PT-2026-59784

Impact The macros that TensorFlow uses for writing assertions e.g., CHECK LT, CHECK GT, etc. have an incorrect logic when comparing size t and int values. Due to type conversion rules, several of the macros would trigger incorrectly. Patches We have patched the issue in GitHub commit...

PT-2026-59949

Impact The implementation of tf.raw ops.StagePeek does not fully validate the input arguments. This results in a CHECK-failure which can be used to trigger a denial of service attack: python import tensorflow as tf index = tf.constant, shape=0, dtype=tf.int32 tf.raw ops.StagePeekindex=index,...

PT-2026-59805

Impact An attacker can craft a TFLite model that would trigger a division by zero in BiasAndClamp implementation: cc inline void BiasAndClampfloat clamp min, float clamp max, int bias size, const float bias data, int array size, float array data // ... TFLITE DCHECK EQarray size % bias size, 0; /...

PT-2026-59972

Impact The implementation of tf.raw ops.QuantizedConv2D does not fully validate the input arguments: python import tensorflow as tf input = tf.constant1, shape=1, 2, 3, 3, dtype=tf.quint8 filter = tf.constant1, shape=1, 2, 3, 3, dtype=tf.quint8 bad args min input = tf.constant, shape=0,...

PT-2026-59798

Impact A malicious user can cause a denial of service by altering a SavedModel such that Grappler optimizer would attempt to build a tensor using a reference dtype. This would result in a crash due to a CHECK-fail in the Tensor constructor as reference types are not allowed. Patches We have patch...

PT-2026-59770

Impact When decoding a tensor from protobuf, TensorFlow might do a null-dereference if attributes of some mutable arguments to some operations are missing from the proto. This is guarded by a DCHECK: cc const auto attr = attrs.Findarg-s; DCHECKattr != nullptr; if attr-value case == AttrValue::kLi...

PT-2026-59732

Impact During shape inference, TensorFlow can allocate a large vector based on a value from a tensor controlled by the user: cc const auto num dims = Valueshape dim; std::vector dims; dims.reservenum dims; Patches We have patched the issue in GitHub commit 1361fb7e29449629e1df94d44e0427ebec8c83c7...

PT-2026-59868

Impact In multiple places, TensorFlow uses tempfile.mktemp to create temporary files. While this is acceptable in testing, in utilities and libraries it is dangerous as a different process can create the file between the check for the filename in mktemp and the actual creation of the file by a...

PT-2026-59716

Impact The implementation of SparseCountSparseOutput is vulnerable to a heap overflow: python import tensorflow as tf import numpy as np tf.raw ops.SparseCountSparseOutput indices=-1,-1, values=2, dense shape=1, 1, weights=1, binary output=True, minlength=-1, maxlength=-1, name=None Patches We ha...

PT-2026-59762

Impact The implementation of StringNGrams can be used to trigger a denial of service attack by causing an OOM condition after an integer overflow: python import tensorflow as tf tf.raw ops.StringNGrams data='123456', data splits=0,1, separator='a'15, ngram widths=, left pad='', right pad='', pad...

PT-2026-59775

Impact The implementation of ThreadPoolHandle can be used to trigger a denial of service attack by allocating too much memory: python import tensorflow as tf y = tf.raw ops.ThreadPoolHandlenum threads=0x60000000,display name='tf' This is because the num threads argument is only checked to not be...

PT-2026-59701

Impact The implementation of Dequantize does not fully validate the value of axis and can result in heap OOB accesses: python import tensorflow as tf @tf.function def test: y = tf.raw ops.Dequantize input=tf.constant1,1,dtype=tf.qint32, min range=1.0, max range=10.0, mode='MIN COMBINED', narrow...

Hugging Face Transformers Regular Expression Denial of Service (ReDoS) vulnerability

A Regular Expression Denial of Service ReDoS vulnerability exists in the Hugging Face Transformers library, specifically in the converttfweightnametoptweightname function. This function, responsible for converting TensorFlow weight names to PyTorch format, uses a regex pattern /^/^// that can be...

TensorFlow has segfault in array_ops.upper_bound

Impactarrayops.upperbound causes a segfault when not given a rank 2 tensor. PatchesWe have patched the issue in GitHub commit 915884fdf5df34aaedd00fc6ace33a2cfdefa586.The fix will be included in TensorFlow 2.13. We will also cherrypick this commit in TensorFlow 2.12.1. For more informationPlease...

PYSEC-2026-1962 TensorFlow has segfault in array_ops.upper_bound

Impact arrayops.upperbound causes a segfault when not given a rank 2 tensor. Patches We have patched the issue in GitHub commit 915884fdf5df34aaedd00fc6ace33a2cfdefa586. The fix will be included in TensorFlow 2.13. We will also cherrypick this commit in TensorFlow 2.12.1. For more information...

Invalid char to bool conversion when printing a tensor

ImpactWhen printing a tensor, we get it's data as a const char array since that's the underlying storage and then we typecast it to the element type. However, conversions from char to bool are undefined if the char is not 0 or 1, so sanitizers/fuzzers will crash. PatchesWe have patched the issue ...

PYSEC-2026-1954 Invalid char to bool conversion when printing a tensor

Impact When printing a tensor, we get it's data as a const char array since that's the underlying storage and then we typecast it to the element type. However, conversions from char to bool are undefined if the char is not 0 or 1, so sanitizers/fuzzers will crash. Patches We have patched the issu...

MLFlow unsafe deserialization

Deserialization of untrusted data can occur in versions of the MLflow platform running version 2.0.0rc0 or newer, enabling a maliciously uploaded Tensorflow model to run arbitrary code on an end user’s system when interacted with...

PYSEC-2026-1654 MLFlow unsafe deserialization

Deserialization of untrusted data can occur in versions of the MLflow platform running version 2.0.0rc0 or newer, enabling a maliciously uploaded Tensorflow model to run arbitrary code on an end user’s system when interacted with...

Authenticated Local Privilege Escalation vulnerability in Intel Optimization for Tensorflow

Improper buffer restrictions in the IntelR Optimization for Tensorflow software before version 2.12 may allow an authenticated user to potentially enable escalation of privilege via local access...