{"id":"CVE-2022-23572","aliases":["BIT-tensorflow-2022-23572","GHSA-rww7-2gpw-fv6j","PYSEC-2022-136","PYSEC-2022-81","PYSEC-2026-3241"],"url":"https://o3.security/vulnerability/CVE-2022-23572","summary":"Crash when type cannot be specialized in Tensorflow","details":"### Impact\nUnder certain scenarios, TensorFlow can fail to specialize a type during [shape inference](https://github.com/tensorflow/tensorflow/blob/a1320ec1eac186da1d03f033109191f715b2b130/tensorflow/core/framework/shape_inference.cc#L168-L174):\n\n```cc\nvoid InferenceContext::PreInputInit(\n    const OpDef& op_def, const std::vector<const Tensor*>& input_tensors,\n    const std::vector<ShapeHandle>& input_tensors_as_shapes) {\n  const auto ret = full_type::SpecializeType(attrs_, op_def);\n  DCHECK(ret.status().ok()) << \"while instantiating types: \" << ret.status();\n  ret_types_ = ret.ValueOrDie();\n  // ... \n}\n```\n\nHowever, `DCHECK` is a no-op in production builds and an assertion failure in debug builds. In the first case execution proceeds to the `ValueOrDie` line. This results in an assertion failure as `ret` contains an error `Status`, not a value. In the second case we also get a crash due to the assertion failure.\n### Patches\nWe have patched the issue in GitHub commit [cb164786dc891ea11d3a900e90367c339305dc7b](https://github.com/tensorflow/tensorflow/commit/cb164786dc891ea11d3a900e90367c339305dc7b).\n\nThe fix will be included in TensorFlow 2.8.0. We will also cherrypick this commit on TensorFlow 2.7.1, and TensorFlow 2.6.3, as these are also affected and still in supported range.\n\n### For more information\nPlease consult [our security guide](https://github.com/tensorflow/tensorflow/blob/master/SECURITY.md) for more information regarding the security model and how to contact us with issues and questions.","published":"2022-02-04T22:32:29Z","modified":"2026-08-12T12:59:30.864952Z","cvss":{"score":6.5,"severity":"MEDIUM","vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H"},"epss":{"score":0.00992,"percentile":0.59141,"asOf":"2026-08-03"},"cisaKev":null,"exploitsKnown":1,"affectedPackages":[{"ecosystem":"PyPI","name":"tensorflow","fixedVersion":"2.5.3"},{"ecosystem":"PyPI","name":"tensorflow","fixedVersion":"2.6.3"},{"ecosystem":"PyPI","name":"tensorflow","fixedVersion":"2.7.1"},{"ecosystem":"PyPI","name":"tensorflow-cpu","fixedVersion":"2.5.3"},{"ecosystem":"PyPI","name":"tensorflow-cpu","fixedVersion":"2.6.3"},{"ecosystem":"PyPI","name":"tensorflow-cpu","fixedVersion":"2.7.1"},{"ecosystem":"PyPI","name":"tensorflow-gpu","fixedVersion":"2.5.3"},{"ecosystem":"PyPI","name":"tensorflow-gpu","fixedVersion":"2.6.3"},{"ecosystem":"PyPI","name":"tensorflow-gpu","fixedVersion":"2.7.1"}],"fix":{"url":"https://github.com/tensorflow/tensorflow/commit/cb164786dc891ea11d3a900e90367c339305dc7b","label":"tensorflow/tensorflow@cb16478"},"references":[{"type":"WEB","url":"https://github.com/tensorflow/tensorflow/blob/a1320ec1eac186da1d03f033109191f715b2b130/tensorflow/core/framework/shape_inference.cc#L168-L174"},{"type":"ADVISORY","url":"https://github.com/CVEProject/cvelistV5/tree/main/cves/2022/23xxx/CVE-2022-23572.json"},{"type":"ADVISORY","url":"https://github.com/tensorflow/tensorflow/security/advisories/GHSA-rww7-2gpw-fv6j"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2022-23572"},{"type":"FIX","url":"https://github.com/tensorflow/tensorflow/commit/cb164786dc891ea11d3a900e90367c339305dc7b"},{"type":"WEB","url":"https://github.com/pypa/advisory-database/tree/main/vulns/tensorflow-cpu/PYSEC-2022-81.yaml"},{"type":"WEB","url":"https://github.com/pypa/advisory-database/tree/main/vulns/tensorflow-gpu/PYSEC-2022-136.yaml"},{"type":"WEB","url":"https://github.com/tensorflow/tensorflow"}],"provenance":{"sources":["OSV.dev","FIRST.org (EPSS)"],"lastVerified":"2026-08-12T12:59:30.864952Z"}}