We have a technical writeup in the works that will be published later this week. The LLM service is currently Continuum and it uses enclaves on GPU as well, encrypting the data away from human eyes. Our service uses attestation to verify that the GPU provider does not have access to the data.