Sheikh Abdur Raheem Ali comments on Defending Against Model Weight Exfiltration Through Inference Verification

[Sheikh Abdur Raheem Ali]

One natural direction is to run the verifier inside a Trusted Execution Environment (TEE), preventing a compromised inference server from tampering with seeds or observing the verification process (there are many startups that do this kind of thing, like tinfoil.sh).

I think that this approach is also taken by workshop labs.