Tag
This paper presents a cloud-native RAG architecture for secure, high-throughput enterprise AI inference on IBM LinuxONE, using Spyre accelerators to achieve sub-two-second latencies and a 20× performance reduction compared to off-platform solutions.
The article compares confidential inference APIs from Privatemode, Tinfoil, NEAR AI, and Chutes, highlighting their security features like end-to-end encryption and trusted execution environments, along with tradeoffs in model selection and verification maturity.
NVIDIA and HPE are expanding their AI factory collaboration with the NVIDIA Vera CPU for agentic AI, the NVIDIA Agent Toolkit for HPE Private Cloud AI, and NVIDIA Confidential Computing across the portfolio, enabling enterprises to move agentic AI into production.
NVIDIA's Confidential Computing, using Blackwell GPUs, is being adopted by Apple to expand its Private Cloud Compute to Google Cloud, enabling secure server-side inference for Apple Intelligence features while maintaining strong privacy guarantees.
Apple announces that its Private Cloud Compute architecture now extends to third-party hardware, specifically Google's servers, using Nvidia, Intel, and Google security technologies to maintain privacy guarantees for advanced AI models like AFM 3 Cloud Pro.
A novel software-based attack misconfigures the Infinity Fabric to break AMD SEV-SNP security guarantees, allowing a malicious hypervisor arbitrary read/write access to confidential virtual machines.
This paper introduces Kettle, an attested build system that generates cryptographically verifiable software provenance using Trusted Execution Environments (TEEs). It aims to eliminate the build infrastructure and operators from the trust surface by binding provenance documents directly to hardware-signed attestation reports.