Trusting What You Cannot See: Auditable Fine-Tuning and Inference for Proprietary AI

תקציר מקורי באנגליתarXiv:2603.07466v2 Announce Type: replace-cross Abstract: Cloud-based infrastructure has become the dominant platform for deploying large models, particularly large language models (LLMs). Fine-tuning and inference are increasingly delegated to cloud providers for simplified deployment and access to proprietary models, yet this creates a fundamental trust gap. Although cryptographic and TEE-based verification approaches exist, prohibitive proving costs and limited TEE memory prevent them from scaling to modern LLMs, leaving clients unable to practically audit these processes. This lack of transparency creates concrete security risks that can silently compromise service integrity. We present AFTUNE, an auditable and verifiable framework that ensures the computational integrity of cloud-base