{"api_version":"1","generated_at":"2026-04-17T01:04:56+00:00","cve":"CVE-2026-1462","urls":{"html":"https://cve.report/CVE-2026-1462","api":"https://cve.report/api/cve/CVE-2026-1462.json","docs":"https://cve.report/api","cve_org":"https://www.cve.org/CVERecord?id=CVE-2026-1462","nvd":"https://nvd.nist.gov/vuln/detail/CVE-2026-1462"},"summary":{"title":"Safe Mode Bypass in keras-team/keras","description":"A vulnerability in the `TFSMLayer` class of the `keras` package, version 3.13.0, allows attacker-controlled TensorFlow SavedModels to be loaded during deserialization of `.keras` models, even when `safe_mode=True`. This bypasses the security guarantees of `safe_mode` and enables arbitrary attacker-controlled code execution during model inference under the victim's privileges. The issue arises due to the unconditional loading of external SavedModels, serialization of attacker-controlled file paths, and the lack of validation in the `from_config()` method.","state":"PUBLISHED","assigner":"@huntr_ai","published_at":"2026-04-13 15:17:18","updated_at":"2026-04-13 15:17:18"},"problem_types":["CWE-502","CWE-502 CWE-502 Deserialization of Untrusted Data"],"metrics":[{"version":"3.0","source":"security@huntr.dev","type":"Secondary","score":"8.8","severity":"HIGH","vector":"CVSS:3.0/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H","data":{"version":"3.0","vectorString":"CVSS:3.0/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H","baseScore":8.8,"baseSeverity":"HIGH","attackVector":"NETWORK","attackComplexity":"LOW","privilegesRequired":"NONE","userInteraction":"REQUIRED","scope":"UNCHANGED","confidentialityImpact":"HIGH","integrityImpact":"HIGH","availabilityImpact":"HIGH"}},{"version":"3.0","source":"CNA","type":"DECLARED","score":"8.8","severity":"HIGH","vector":"CVSS:3.0/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H","data":{"attackComplexity":"LOW","attackVector":"NETWORK","availabilityImpact":"HIGH","baseScore":8.8,"baseSeverity":"HIGH","confidentialityImpact":"HIGH","integrityImpact":"HIGH","privilegesRequired":"NONE","scope":"UNCHANGED","userInteraction":"REQUIRED","vectorString":"CVSS:3.0/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H","version":"3.0"}}],"references":[{"url":"https://huntr.com/bounties/7e78d6f1-6977-4300-b595-e81bdbda331c","name":"https://huntr.com/bounties/7e78d6f1-6977-4300-b595-e81bdbda331c","refsource":"security@huntr.dev","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://github.com/keras-team/keras/commit/b6773d3decaef1b05d8e794458e148cb362f163f","name":"https://github.com/keras-team/keras/commit/b6773d3decaef1b05d8e794458e148cb362f163f","refsource":"security@huntr.dev","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://www.cve.org/CVERecord?id=CVE-2026-1462","name":"CVE Program record","refsource":"CVE.ORG","tags":["canonical"]},{"url":"https://nvd.nist.gov/vuln/detail/CVE-2026-1462","name":"NVD vulnerability detail","refsource":"NVD","tags":["canonical","analysis"]}],"affected":[{"source":"CNA","vendor":"keras-team","product":"keras-team/keras","version":"affected unspecified 3.13.2 custom","platforms":[]}],"timeline":[],"solutions":[],"workarounds":[],"exploits":[],"credits":[],"nvd_cpes":[],"vendor_comments":[],"enrichments":{"kev":null,"epss":{"cve_year":"2026","cve_id":"1462","cve":"CVE-2026-1462","epss":"0.000600000","percentile":"0.187810000","score_date":"2026-04-15","updated_at":"2026-04-16 00:13:56"},"legacy_qids":[]},"source_records":{"cve_program":{"containers":{"cna":{"affected":[{"product":"keras-team/keras","vendor":"keras-team","versions":[{"lessThan":"3.13.2","status":"affected","version":"unspecified","versionType":"custom"}]}],"descriptions":[{"lang":"en","value":"A vulnerability in the `TFSMLayer` class of the `keras` package, version 3.13.0, allows attacker-controlled TensorFlow SavedModels to be loaded during deserialization of `.keras` models, even when `safe_mode=True`. This bypasses the security guarantees of `safe_mode` and enables arbitrary attacker-controlled code execution during model inference under the victim's privileges. The issue arises due to the unconditional loading of external SavedModels, serialization of attacker-controlled file paths, and the lack of validation in the `from_config()` method."}],"metrics":[{"cvssV3_0":{"attackComplexity":"LOW","attackVector":"NETWORK","availabilityImpact":"HIGH","baseScore":8.8,"baseSeverity":"HIGH","confidentialityImpact":"HIGH","integrityImpact":"HIGH","privilegesRequired":"NONE","scope":"UNCHANGED","userInteraction":"REQUIRED","vectorString":"CVSS:3.0/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H","version":"3.0"}}],"problemTypes":[{"descriptions":[{"cweId":"CWE-502","description":"CWE-502 Deserialization of Untrusted Data","lang":"en","type":"CWE"}]}],"providerMetadata":{"dateUpdated":"2026-04-13T14:55:28.649Z","orgId":"c09c270a-b464-47c1-9133-acb35b22c19a","shortName":"@huntr_ai"},"references":[{"url":"https://huntr.com/bounties/7e78d6f1-6977-4300-b595-e81bdbda331c"},{"url":"https://github.com/keras-team/keras/commit/b6773d3decaef1b05d8e794458e148cb362f163f"}],"source":{"advisory":"7e78d6f1-6977-4300-b595-e81bdbda331c","discovery":"EXTERNAL"},"title":"Safe Mode Bypass in keras-team/keras"}},"cveMetadata":{"assignerOrgId":"c09c270a-b464-47c1-9133-acb35b22c19a","assignerShortName":"@huntr_ai","cveId":"CVE-2026-1462","datePublished":"2026-04-13T14:55:28.649Z","dateReserved":"2026-01-27T04:14:51.848Z","dateUpdated":"2026-04-13T14:55:28.649Z","state":"PUBLISHED"},"dataType":"CVE_RECORD","dataVersion":"5.2"},"nvd":{"publishedDate":"2026-04-13 15:17:18","lastModifiedDate":"2026-04-13 15:17:18","problem_types":["CWE-502","CWE-502 CWE-502 Deserialization of Untrusted Data"],"metrics":{"cvssMetricV30":[{"source":"security@huntr.dev","type":"Secondary","cvssData":{"version":"3.0","vectorString":"CVSS:3.0/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H","baseScore":8.8,"baseSeverity":"HIGH","attackVector":"NETWORK","attackComplexity":"LOW","privilegesRequired":"NONE","userInteraction":"REQUIRED","scope":"UNCHANGED","confidentialityImpact":"HIGH","integrityImpact":"HIGH","availabilityImpact":"HIGH"},"exploitabilityScore":2.8,"impactScore":5.9}]},"configurations":[]},"legacy_mitre":{"record":{"CveYear":"2026","CveId":"1462","Ordinal":"1","Title":"Safe Mode Bypass in keras-team/keras","CVE":"CVE-2026-1462","Year":"2026"},"notes":[{"CveYear":"2026","CveId":"1462","Ordinal":"1","NoteData":"A vulnerability in the `TFSMLayer` class of the `keras` package, version 3.13.0, allows attacker-controlled TensorFlow SavedModels to be loaded during deserialization of `.keras` models, even when `safe_mode=True`. This bypasses the security guarantees of `safe_mode` and enables arbitrary attacker-controlled code execution during model inference under the victim's privileges. The issue arises due to the unconditional loading of external SavedModels, serialization of attacker-controlled file paths, and the lack of validation in the `from_config()` method.","Type":"Description","Title":"Safe Mode Bypass in keras-team/keras"}]}}}