AI Sec Watch

TensorFlow, an open source machine learning platform, has a vulnerability in its `tf.function` API (a feature that converts Python functions into optimized operations) where mutually recursive functions (functions that call each other back and forth) can cause a deadlock using a non-reentrant Lock (a mechanism that prevents simultaneous access but doesn't allow the same thread to re-enter it). An attacker could cause a denial of service by tricking users into loading vulnerable models, though this scenario is uncommon.

TensorFlow, an open source machine learning platform, has a bug in its shape inference code for the `AllToAll` function that causes a division by zero error (when a value is divided by 0, causing the program to crash) whenever the `split_count` argument is set to 0. This vulnerability could allow an attacker to crash or disrupt a TensorFlow application.

TensorFlow (an open source platform for machine learning) has a bug where its convolution operators (mathematical functions that process data in neural networks) crash with a division by zero error when given empty filter tensors (arrays of parameters). This vulnerability affects multiple versions of TensorFlow.

TensorFlow's boosted trees code (a machine learning feature for building multiple decision trees together) lacks proper input validation, allowing attackers to crash the system (denial of service, where a service becomes unavailable), read sensitive data from memory, or write malicious data to memory buffers. The TensorFlow developers recommend stopping use of these APIs since the boosted trees code is no longer actively maintained.

TensorFlow, an open source platform for machine learning, has a vulnerability in its `ParallelConcat` function that lacks proper input validation and can cause a division by zero error (a crash caused by dividing a number by zero). The affected versions have known fixes available through updates to TensorFlow 2.7.0 and earlier supported versions.

TensorFlow, a machine learning platform, has a vulnerability (CVE-2021-41206) where certain operations don't properly check the size and dimensions of tensor arguments (the numerical arrays that machine learning models process). This missing validation can cause crashes, memory corruption (reads and writes to unintended memory locations), or other undefined behavior depending on which operation is affected.

TensorFlow, an open source platform for machine learning, has a bug in its `tf.range` function where a conditional statement mixes two different number types (int64, a large integer type, and double, a decimal number type). Due to how C++ automatically converts between these types, the calculation overflows (produces incorrect results that are too large to store). This causes the output size calculation to fail.

TensorFlow, an open source platform for machine learning, has a vulnerability in its `SparseBinCount` function that allows heap OOB access (out-of-bounds memory access, where a program reads data outside the memory it's allowed to use) because it doesn't validate that the `values` argument matches the shape of the sparse output. This bug could let attackers crash the system or potentially read sensitive data from memory.

TensorFlow, an open source machine learning platform, has a vulnerability in the `SparseFillEmptyRows` function that can cause a heap OOB access (out-of-bounds read, where a program tries to read memory it shouldn't access) when the size of `indices` does not match the size of `values`. This is a memory safety bug that could potentially crash the program or expose sensitive data.