CVE-2026-40165: authentik SAML Authentication Bypass

Are You Affected?

CVE-2026-40165 affects authentik deployments that process signed SAML assertions through a configured SAML Source. It does not affect every authentik installation. Exploitation requires:

• An affected authentik version
• A configured SAML Source
• XML signing enabled for the relevant assertion
• An attacker-controlled account at the upstream SAML identity provider
• The ability for that attacker to influence the SAML NameID value

Start by inventorying authentik instances and their enabled sources. Prioritize instances where SAML sources have XML signing enabled. Then determine whether the upstream identity provider allows users to alter usernames, email addresses, subject identifiers, or other attributes that become the SAML NameID.

The impact is potentially account takeover at the authentik layer. A successful attacker could cause authentik to map the manipulated identity to another existing account, including a privileged administrator account.

Organizations managing multiple identity systems should also review their vulnerability assessment process and identity-provider inventory to ensure affected authentik deployments are included.

The Fix, If You’re in a Hurry

Upgrade authentik to 2025.12.5 or later when running the 2025.12 branch. Deployments on the 2026.2 release line should upgrade to 2026.2.3 or later.

The fixed versions are:

• 2025.12.5
• 2026.2.3

For a Docker Compose deployment, update the authentik image tag in the compose configuration and recreate the services:

# Replace the image tag with the fixed release appropriate to your branch.
# Confirm the exact image repository and deployment procedure in your environment.
docker compose pull
docker compose up -d
docker compose ps

Pulling an image alone does not change running containers. Confirm the running version after the rollout and verify that all authentik worker and server components use the fixed release.

For Kubernetes, update the image tag in the deployment or Helm values, apply the change, and verify rollout status using your organization’s approved authentik deployment procedure.

If immediate upgrading is impossible, apply the workaround documented in the authentik security advisory. The available research confirms that a vendor-documented workaround exists but does not reproduce its exact configuration steps. Do not invent or substitute configuration changes. Consult the advisory and test the workaround against a staging SAML flow before applying it to production.

As an interim control, consider disabling affected SAML sources or preventing untrusted users from changing the upstream attribute that becomes NameID.

After remediation, invalidate sessions and require reauthentication for users who authenticated through potentially affected SAML sources if you cannot rule out prior exploitation. Review privileged accounts first.

Analyst’s Take: Patch based on configuration and account impact, not on the absence of a KEV listing or public proof of concept. The highest-priority checks are affected authentik versions, signed SAML sources, upstream control of NameID, and privileged accounts that used those flows.

How This Vulnerability Works

The vulnerability is rooted in how authentik extracted the SAML NameID value from an XML assertion. In the affected implementation, extraction could return only the text preceding the first child node rather than the complete text content of the element. XML comments are child nodes, so an attacker could use a comment to split the apparent identity value.

Conceptually, an upstream identity value could be structured so that the text before an XML comment matched another authentik account, while the remaining text preserved the attacker’s original value or otherwise satisfied upstream processing. When authentik mapped the truncated value, it could interpret the assertion as belonging to the account represented by the prefix.

The attack requires more than arbitrary internet access. The attacker needs:

• An account at the upstream SAML source
• The ability to influence the resulting NameID
• A configured authentik SAML Source
• XML signing enabled for the assertion

These conditions can exist in enterprise federations where the upstream IdP allows users or tenants to manage identity attributes.

The fix changes extraction to use the complete concatenated text content of the NameID element rather than only the text preceding the first child node. The referenced upstream fix commit is:

47dec5c6b7fb4a62bfad2ae8bddf002bde7ba774

Administrators should validate complete SAML mapping behavior after upgrading, especially where custom attribute mappings, account linking, or privileged-group mappings are configured.

Technical Notes

The vulnerable behavior concerns XML text-node handling. A simplified representation of the problematic identity shape is:

<saml:NameID>target-account<!-- attacker-controlled remainder -->@example.invalid</saml:NameID>

This is illustrative of the parsing issue, not a production exploit recipe. The security-relevant point is that the parser and application must use the full text content, not only the first text node.

Severity, Explained

The reported CVSS base score is 8.7, placing the issue in the high-severity category. The supplied NVD response did not expose the complete CVSS vector string, so the vector should be confirmed directly in the NVD record before publishing a component-by-component score breakdown.

Operationally, the high score reflects the potential security consequence: successful exploitation can bypass normal identity mapping and authenticate an attacker as another user.

The likely impact is strongest where authentik protects:

• Administrative consoles
• VPN access
• Internal applications
• Cloud control planes
• Other systems that trust authentik as an identity provider

The prerequisites reduce exposure compared with an unauthenticated remote attack. The attacker must possess an upstream SAML account and control a value used as NameID; the target must also use an affected authentik release and matching SAML configuration.

Those requirements are not sufficient reason to defer remediation. In many organizations, upstream accounts are broadly available to employees, contractors, customers, or tenants, and a successful identity substitution could affect high-value targets.

Because the exact vector is not available in the supplied record, security teams should preserve the published score of 8.7 while verifying the current NVD entry and vendor advisory during risk acceptance, vulnerability prioritization, and compliance reporting.

Is It Being Exploited?

CVE-2026-40165 is not currently listed in CISA’s Known Exploited Vulnerabilities catalog. The supplied assessment reports on_kev: false, with no CISA date-added entry, federal remediation deadline, required action, or ransomware campaign flag.

No confirmed in-the-wild exploitation was identified in the reviewed NVD, CISA KEV, authentik advisory, or search results. No public proof-of-concept repository was identified in the reviewed sources either.

The current status is technically exploitable under the stated conditions, but neither active exploitation nor a public proof of concept is confirmed.

This status does not mean exploitation is impossible. Authentication bypass vulnerabilities in identity providers can be attractive targets, and exploitation may leave limited traces if an attacker successfully receives a normal authenticated session. Organizations should patch based on exposure and impact rather than waiting for a KEV listing or public exploit.

The supplied research note records one GitHub security advisory and the upstream fix commit, but those references are remediation and disclosure artifacts, not evidence of an operational exploit.