CVE-2026-119344: Cocos Intra-handshake attested TLS implementation is vulnerable to Diversion Attacks
Cocos AI is a confidential computing system for AI. In version 0.8.2 and earlier, Cocos AI accepts trusted execution environment attestation Evidence when firmware patch levels and software measurements match, without binding the Evidence to an allowed platform owner or unique machine identity and without enforcing revocation. A remote attacker who controls or compromises another platform trusted by the same verifier can substitute valid Evidence from that platform and cause authorization of the wrong platform. This vulnerability is addressed in 0.9.0. However, this vulnerability can't be patched in source code alone, since the vulnerability arises from the confidential computing threat model itself.
Remediation
Recommended actions to resolve this vulnerability, in priority order.
- Upgrade
Upgrade
Cocos AIto a version that resolves this vulnerability.Fixed in 0.9.0
Event History
Frequently Asked Questions
Which deployments are exposed to this issue?
Cocos AI deployments running version 0.8.2 or earlier are affected when their verifier accepts TEE attestation Evidence based on matching firmware patch levels and software measurements without also binding it to an allowed platform owner or unique machine identity and enforcing revocation.
What does an attacker need to exploit it?
The attacker must control or compromise another platform that is trusted by the same verifier. They can then provide valid Evidence from that platform to cause authorization of the wrong platform.
Is upgrading to version 0.9.0 sufficient to eliminate the risk?
Version 0.9.0 addresses the vulnerability, but the issue cannot be patched in source code alone because it arises from the confidential-computing threat model. Deployments also need controls that bind Evidence to an allowed platform owner or unique machine identity and enforce revocation.