CVE-2020-5215: TensorFlow: type confusion DoS crashes eager mode inference

In TensorFlow before 1.15.2 and 2.0.1, converting a string (from Python) to a tf.float16 value results in a segmentation fault in eager mode as the format checks for this use case are only in the graph mode. This issue can lead to denial of service in inference/training where a malicious attacker can send a data point which contains a string instead of a tf.float16 value. Similar effects can be obtained by manipulating saved models and checkpoints whereby replacing a scalar tf.float16 value with a scalar string will trigger this issue due to automatic conversions. This can be easily reproduced by tf.constant("hello", tf.float16), if eager execution is enabled. This issue is patched in TensorFlow 1.15.1 and 2.0.1 with this vulnerability patched. TensorFlow 2.1.0 was released after we fixed the issue, thus it is not affected. Users are encouraged to switch to TensorFlow 1.15.1, 2.0.1 or 2.1.0.

An adversary targeting an AI-powered API (e.g., a fraud detection or NLP preprocessing service backed by TensorFlow) identifies that the inference endpoint accepts JSON payloads with tensor data. The attacker submits a request with a string value ('hello' or any non-numeric string) in a field expected to be a float16 tensor. TensorFlow's eager execution attempts the type conversion, hits the unguarded code path, and the process segfaults. With a scripted loop, the attacker can maintain a persistent DoS against the service. Alternatively, for a more targeted supply chain attack, the adversary compromises a shared model registry or CI artifact store and replaces a scalar float16 value in a saved checkpoint with a string — every deployment that loads this checkpoint crashes on startup, causing a service outage affecting all consumers of that model artifact.