AMD and Google Cloud Put Confidential AI Inside the TPU Path

Google Cloud is bringing hardware-backed confidential computing to TPU-based AI, protecting sensitive workloads across the host and accelerator path.

Updated

What is this trend?

Google Cloud is extending confidential computing from CPUs into TPU-based AI, using AMD EPYC and SEV/Infinity Guard to protect prompts, weights, and data during inference and training.

  • First cloud TPU deployment with hardware-enforced confidential computing
  • Protects sensitive AI data across the EPYC host and TPU path
  • Extends trusted execution beyond CPU-only confidential AI
  • Raises the bar for secure inference, fine-tuning, and RL on shared cloud
  • Signals a broader race to make attestation and key release default AI controls

What’s the latest?

On November 9, AMD and Google Cloud pushed the story further: Google Cloud’s new TPU 8tx runs on 5th-generation AMD EPYC processors with AMD SEV/Infinity Guard, and Google says it is the first cloud T

How it developed

  1. AI Governance Becomes the New Control Plane, Confidential Computing Enters Kubernetes Control Plane
    • Confidential Computing Moves Into the Kubernetes Control Plane

Go deeper

Curated long-form picks on this trend — podcasts, videos, and analysis, by vantage.

Stay ahead in Cloud Computing

Get the weekly brief in your inbox — the developments, what they mean by vantage, and what to do next.