CAI Content Credentials EUVD-2026-100737

Exploitation requires that a target application or service actually invoke the CAI Content Credentials SDK to verify the cryptographic signature of a Content Credentials (C2PA-style) manifest, and that the attacker can supply the asset/manifest to that verification path; the flaw lets a forged or tampered signature be accepted as valid, granting unauthorized write access to credential data. Per the CVSS vector (AV:N/AC:L/PR:N/UI:N), no authentication and no user interaction are needed, so the primary limiting factors are deployment-specific: the vulnerable SDK version must be in use and the application must rely on its signature verification as a trust boundary. The impact is confined to integrity (C:N/A:N), so confidentiality and availability are not affected, and the practical damage depends on how much the consuming system trusts verified credentials.

This is a moderate-priority integrity issue rather than a top-tier emergency. The provided CVSS 3.1 vector (AV:N/AC:L/PR:N/UI:N) describes a network-reachable, low-complexity, unauthenticated bypass with high integrity impact but no confidentiality or availability impact (C:N/I:H/A:N), yielding a 7.5 base score consistent with a signature-verification weakness (CWE-347) that lets an attacker forge or tamper with content credentials. Two key signals are MISSING: there is no EPSS exploitation-probability score and no SSVC decision data in the input, so automatability and real-world exploitation likelihood cannot be quantified. The CVE is NOT listed in CISA KEV and no public POC is indicated, so there is no evidence of active exploitation - the only exploitation enabler on record is the low attack complexity. The 'network' attack vector should be interpreted carefully: a cryptographic-signature-verification bypass is typically exploited by presenting a crafted/forged signed asset to a verifier, so 'network' here reflects that the malicious input can arrive remotely, not that a listening service is being attacked - verify the exact exploitation surface with Adobe's advisory. Net: a real but bounded integrity/trust-model flaw, meaningful for anyone relying on Content Credentials for provenance authenticity, but not a drop-everything issue absent exploitation evidence.

An attacker crafts a media asset bearing a forged or improperly-signed Content Credentials manifest and submits it to an application that uses the CAI Content Credentials SDK to verify provenance. Because the SDK improperly verifies the cryptographic signature, the tampered manifest is accepted as authentic, letting the attacker gain unauthorized write access to credential data and pass off manipulated content as genuine. The network vector and low complexity (AV:N/AC:L) with no required authentication or user interaction make the forgery straightforward to deliver, though no public exploit code is currently identified.