Secure Enclave Inference

AI inference occurs in a confidential off-chain environment (e.g., SGX or MPC).

🔄 Flow:

• Enclave decrypts the input

• Runs neural model (e.g., PyTorch or ONNX)

• Produces an output prediction

• Keeps both input & model parameters private

🛡️ Ensures sensitive data never leaves the enclave.