TEE Attestation Verifier

Every InjectMe judgment is produced inside a hardware Trusted Execution Environment on 0G Compute and archived on 0G Storage. Enter a Chat ID to verify the cryptographic proof of inference.

Chat IDs are returned when you submit an attack on any challenge.

How TEE Verification Works

InjectMe runs all AI inference inside 0G Compute Trusted Execution Environments. Every judgment is cryptographically attested by the hardware, not just signed by a server key.

01Sealed Inference

Each AI inference runs inside a hardware-sealed TEE on 0G Compute nodes. The model weights and system prompt are loaded into encrypted memory that the host operator cannot read or modify.

02Remote Attestation

After inference, the TEE generates a cryptographic attestation report signed by the CPU hardware. This report binds the exact code hash, input, and output to the hardware environment.

03Storage Archival

The judgment, conversation hash, and attestation proof are stored on 0G Storage. A Merkle root hash anchors the full record, making any post-hoc tampering detectable.

04Live Re-verification

At query time, the backend re-checks the attestation against the 0G Compute attestation registry. If the hardware certificate has been revoked or the record altered, live verification fails.