Curious if ypu will be providing any insight in future write ups on the technology about the GPU side of things. Whose GPU? Where is it? How is it secured?
Also, (I'm still learning how a lot of this stuff works so this may be a naive question) doesn't the GPU need to see the info in order to provide an AI response?
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.
Excellent. Thanks for the response. I'll look out for the writeup.
