CVE-2025-12058

CVE-2025-12058 is a medium-severity vulnerability with a CVSS 4.0 base score of 5.9. It is not currently listed as actively exploited by CISA, and its EPSS exploit-prediction score is low. The underlying weakness is classified as CWE-502.

Key facts

Description

The Keras.Model.load_model method, including when executed with the intended security mitigation safe_mode=True, is vulnerable to arbitrary local file loading and Server-Side Request Forgery (SSRF). This vulnerability stems from the way the StringLookup layer is handled during model loading from a specially crafted .keras archive. The constructor for the StringLookup layer accepts a vocabulary argument that can specify a local file path or a remote file path. * Arbitrary Local File Read: An attacker can create a malicious .keras file that embeds a local path in the StringLookup layer's configuration. When the model is loaded, Keras will attempt to read the content of the specified local file and incorporate it into the model state (e.g., retrievable via get_vocabulary()), allowing an attacker to read arbitrary local files on the hosting system. * Server-Side Request Forgery (SSRF): Keras utilizes tf.io.gfile for file operations. Since tf.io.gfile supports remote filesystem handlers (such as GCS and HDFS) and HTTP/HTTPS protocols, the same mechanism can be leveraged to fetch content from arbitrary network endpoints on the server's behalf, resulting in an SSRF condition. The security issue is that the feature allowing external path loading was not properly restricted by the safe_mode=True flag, which was intended to prevent such unintended data access.

Frequently asked questions

What is CVE-2025-12058?
The Keras.Model.load_model method, including when executed with the intended security mitigation safe_mode=True, is vulnerable to arbitrary local file loading and Server-Side Request Forgery (SSRF). This vulnerability stems from the way the StringLookup layer is handled during model loading from a specially crafted .keras archive. The constructor for the StringLookup layer accepts a vocabulary argument that can specify a local file path or a remote file path. * Arbitrary Local File Read: An attacker can create a malicious .keras file that embeds a local path in the StringLookup layer's configuration. When the model is loaded, Keras will attempt to read the content of the specified local file and incorporate it into the model state (e.g., retrievable via get_vocabulary()), allowing an attacker to read arbitrary local files on the hosting system. * Server-Side Request Forgery (SSRF): Keras utilizes tf.io.gfile for file operations. Since tf.io.gfile supports remote filesystem handlers (such as GCS and HDFS) and HTTP/HTTPS protocols, the same mechanism can be leveraged to fetch content from arbitrary network endpoints on the server's behalf, resulting in an SSRF condition. The security issue is that the feature allowing external path loading was not properly restricted by the safe_mode=True flag, which was intended to prevent such unintended data access.
How severe is CVE-2025-12058?
CVE-2025-12058 has a CVSS 4.0 base score of 5.9, rated medium severity.
Is CVE-2025-12058 being actively exploited?
It is not currently listed in CISA's KEV catalog. Its EPSS exploit-prediction score is 0% (15th percentile), an estimate of the probability of exploitation in the next 30 days.
How do I fix CVE-2025-12058?
Review the linked vendor and NVD advisories for patched versions and mitigations, then upgrade or apply the recommended workaround.
Does CVE-2025-12058 have an EU (EUVD) identifier?
Yes. CVE-2025-12058 is tracked in the ENISA EU Vulnerability Database (EUVD) as EUVD-2025-36634.
When was CVE-2025-12058 published?
CVE-2025-12058 was published on 2025-10-29 and last updated on 2026-06-17.

References

Other CWE-502 (Deserialization of Untrusted Data) vulnerabilities

Browse all CWE-502 (Deserialization of Untrusted Data) vulnerabilities →