{"id":"CVE-2026-44843","title":"langchain: LangChain: Information disclosure and data integrity compromise via insecure deserialization (CVE-2026-44843)","summary":"A flaw was found in LangChain, a framework for building agents and large language model (LLM)-powered applications. Older runtime code paths within LangChain deserialize application-controlled payloads using overly broad object allowlists.…","severity":"high","cvss":8.2,"cvssVector":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:L/A:N","cvssSource":"vendor","cwe":"CWE-502","vendor":"Red Hat","product":"Red Hat OpenShift AI 3.4","affected":["exploit_intelligence","openshift_lightspeed","ansible_automation_platform 2","enterprise_linux_ai_rhel_ai 3","migration_toolkit_for_applications 8.2","openshift_ai 3.3","openshift_ai 3.4"],"patched":["migration_toolkit_for_applications 8.2","openshift_ai 3.3","openshift_ai 3.4"],"published":"2026-05-26","updated":"2026-10-19","sourceUpdated":"2026-10-19T11:17:22+00:00","source":"CSAF","sourceUrl":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-44843.json","references":[{"url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-44843.json"},{"url":"https://access.redhat.com/security/cve/CVE-2026-44843"},{"url":"https://bugzilla.redhat.com/show_bug.cgi?id=2481695"},{"url":"https://www.cve.org/CVERecord?id=CVE-2026-44843"},{"url":"https://nvd.nist.gov/vuln/detail/CVE-2026-44843"},{"url":"https://github.com/langchain-ai/langchain/security/advisories/GHSA-pjwx-r37v-7724"},{"url":"https://access.redhat.com/errata/RHSA-2026:43038"},{"url":"https://access.redhat.com/errata/RHSA-2026:73987"},{"url":"https://access.redhat.com/errata/RHSA-2026:60520"},{"url":"https://access.redhat.com/errata/RHSA-2026:84734"},{"url":"https://github.com/langchain-ai/langchain"}],"tags":["csaf","vex","red-hat","osv","pip"],"epss":0.00406,"epssPercentile":0.34698,"aliases":["GHSA-pjwx-r37v-7724","PYSEC-2026-2564"],"ecosystem":"pip","ingestedAt":"2026-07-13T18:58:01.543Z","slug":"CVE-2026-44843","body":"## Overview\n\nA flaw was found in LangChain, a framework for building agents and large language model (LLM)-powered applications. Older runtime code paths within LangChain deserialize application-controlled payloads using overly broad object allowlists. A remote attacker can exploit this by supplying specially crafted LangChain serialized constructor dictionaries, leading to the instantiation of classes with untrusted arguments. This can result in the disclosure of sensitive information and potential partial compromise of data integrity.\n\n## Vendor advisories\n\n- **RHSA-2026:43038** · Red Hat · fixed in: Red Hat Migration Toolkit for Applications 8.2 · released 2026-07-21 · [advisory](https://access.redhat.com/errata/RHSA-2026:43038)\n- **RHSA-2026:73987** · Red Hat · fixed in: Red Hat OpenShift AI 3.3 · released 2026-09-30 · [advisory](https://access.redhat.com/errata/RHSA-2026:73987)\n- **RHSA-2026:60520** · Red Hat · fixed in: Red Hat OpenShift AI 3.4 · released 2026-08-27 · [advisory](https://access.redhat.com/errata/RHSA-2026:60520)\n- **RHSA-2026:84734** · Red Hat · fixed in: Red Hat OpenShift AI 3.4 · released 2026-10-19 · [advisory](https://access.redhat.com/errata/RHSA-2026:84734)\n- **Red Hat VEX** · Important · affected: Exploit Intelligence, OpenShift Lightspeed, Red Hat Ansible Automation Platform 2, Red Hat Enterprise Linux AI (RHEL AI) 3 · no fix planned: Red Hat Ansible Automation Platform 2, Red Hat Enterprise Linux AI (RHEL AI) 3, Exploit Intelligence, OpenShift Lightspeed · updated 2026-10-19 · [vex](https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-44843.json)\n\n**langchain: LangChain: Information disclosure and data integrity compromise via insecure deserialization** — rated Important by Red Hat. Released 2026-05-26, updated 2026-10-19.\n\nAffected:\n\n- Exploit Intelligence\n- OpenShift Lightspeed\n- Red Hat Ansible Automation Platform 2\n- Red Hat Enterprise Linux AI (RHEL AI) 3\n\nFixed:\n\n- Red Hat Migration Toolkit for Applications 8.2\n- Red Hat OpenShift AI 3.3\n- Red Hat OpenShift AI 3.4\n\nNo fix planned:\n\n- Red Hat Ansible Automation Platform 2\n- Red Hat Enterprise Linux AI (RHEL AI) 3\n- Exploit Intelligence\n- OpenShift Lightspeed\n\nNot affected:\n\n- Red Hat Migration Toolkit for Applications 8.2\n- Red Hat OpenShift AI 3.3\n- Red Hat OpenShift AI 3.4\n\n## Remediation\n\nBefore applying this update, make sure all previously released errata\nrelevant to your system have been applied. https://access.redhat.com/errata/RHSA-2026:43038\nFor Red Hat OpenShift AI 3.3.7 see the following documentation, which will be updated shortly for this release, for important instructions on how to upgrade your cluster and fully apply this errata update:\n\nhttps://docs.redhat.com/en/documentation/red_hat_openshift_ai/ https://access.redhat.com/errata/RHSA-2026:73987\nFor Red Hat OpenShift AI 3.4.4 see the following documentation, which will be updated shortly for this release, for important instructions on how to upgrade your cluster and fully apply this errata update:\n\nhttps://docs.redhat.com/en/documentation/red_hat_openshift_ai/ https://access.redhat.com/errata/RHSA-2026:60520\n\nWorkarounds / mitigations:\n\n- To mitigate this issue, applications using LangChain should avoid passing untrusted structured input directly to load() or loads() functions. Instead, ensure all incoming untrusted data is validated and canonicalized into an inert schema before being processed by LangChain. Additionally, migrate away from deprecated LangChain APIs, including RunnableWithMessageHistory, astream_log(), and astream_events(version=\"v1\"), as these older code paths are more susceptible to insecure deserialization.\n\n## Package advisory (CVE-2026-44843)\n\nAffected packages:\n\n- `langchain-core >= 1.0.0, < 1.3.3`\n- `langchain-core < 0.3.85`\n\nPatched in:\n\n- `langchain-core 1.3.3`\n- `langchain-core 0.3.85`\n\nSource: https://osv.dev/vulnerability/GHSA-pjwx-r37v-7724","depth":"twilight","depthScore":45,"depthScoreParts":{"impact":45.1,"likelihood":0.1,"exploitation":0,"ransomware":0},"changes":[]}