CVE-2021-29545

CVE-2021-29545 is a denial-of-service vulnerability affecting TensorFlow, specifically versions prior to 2.5.0, 2.4.2, 2.3.3, 2.2.3, and 2.1.4. It stems from a boundary error during the conversion of sparse tensors to CSR Sparse matrices, where an attacker can trigger a CHECK-fail by providing malformed input that causes an out-of-bounds write on heap-allocated data. Rated as Medium severity (CVSS 5.5), this vulnerability requires local access and low privileges (AV:L/PR:L) to exploit, with no user interaction needed. A successful attack leads to a high impact on availability (A:H), resulting in a denial of service. There is currently no evidence of active exploitation, nor are there publicly available exploit tools like Metasploit, Nuclei, or ExploitDB modules. Community discussion and media coverage for this CVE are minimal, indicating low public awareness.