Scaling Trust: The Fortanix and Rafay Integration for Enterprise Confidential AI

In the modern enterprise, Artificial Intelligence (AI) has moved from a "nice-to-have" experimental phase to a core business driver. However, for organizations in highly regulated sectors—such as banking, healthcare, and government—the path to AI adoption is fraught with security hurdles.

The primary concern is protecting sensitive data not just at rest or in transit, but in use. In the image below, the app uses a proprietary model which needs to be secured using confidential computing.

Traditional security measures often fall short when data must be decrypted to be processed by an AI model. This is where Confidential Computing changes the game, and why the joint integration between Fortanix and Rafay is a landmark development for the "AI Factory" of the future.

The Vision: Secure and Sovereign AI

Consider a financial institution operating in a strictly regulated environment. They aim to be at the forefront of technology by deploying Speech-to-Text services (e.g. based on NVIDIA's Parakeet NIM) to boost internal productivity.

Because of the need for rapid deployments, it is common to want to partner with neo-clouds that can provide them with access to cutting edge GPU and associated AI infrastructure. However, the financial institution's compliance mandates are non-negotiable:

• Zero Trust Architecture: Security must be ingrained at every layer, assuming no implicit trust.
• Data Sovereignty: They must maintain end-to-end control within their own cloud environment.
• Operational Scale: The solution cannot be a one-off; it must be repeatable and automated for various AI use cases.

The Powerhouse Partnership: Fortanix + Rafay

The integration combines Fortanix’s industry-leading data security with Rafay’s robust orchestration platform to create a turnkey solution for Confidential AI.

Fortanix: The Security Foundation

Fortanix provides the "vault" for AI workloads through its Confidential Computing Manager (CCM) and Data Security Manager (DSM).

Confidential Computing Manager (CCM)

This module creates and manages Confidential Virtual Machines (CVMs) and containers. It performs Composite Attestation, a process that verifies the integrity of the hardware, the BIOS, and the software stack before any sensitive data is processed.

Data Security Manager (DSM)

This acts as a hardened Key Management System (KMS) and Hardware Security Module (HSM). It ensures that encryption keys are only released to the AI workload after the CCM has validated the attestation.

Rafay: The Orchestration Engine

While Fortanix secures the what, Rafay handles the how. Rafay provides the automation and deployment scale required to manage these secure environments across distributed infrastructure.

Repeatable Workflows

Enable administrators to define "SKUs" for the application, the underlying form factor (VM, bare metal server, Kubernetes etc) and all necessary security configurations.

Centralized Management, Orchestration, Self Service

Through the Rafay platform, neoclouds can manage multiple tenants, manage data centers with heterogenous AI infrastructure, provide end users with a self service portal to launch and use the SKUs.

Anatomy of a Confidential AI Deployment

The true value of this integration is best seen through the simplified workflow it provides for different organizational personas.

1. The Blueprint (SKU Creator)

The workflow begins with the SKU Creator. This persona defines the "blueprint" for the Speech-to-Text application/service. They work with the neo-cloud admin to ensure the environment has the correct NVIDIA GPUs (such as H100 or Blackwell) capable of supporting hardware-based Trusted Execution Environments (TEEs).

2. The Infrastructure (Neo Cloud Admin)

The Infrastructure Admin at the neo-cloud vendor uses Rafay to provision the underlying hardware and hypervisor. They ensure that the Fortanix node agents are in place to facilitate communication between the hardware powered secure enclaves and the Fortanix management plane.

3. The One-Click Experience (End User)

For the developer or end user, the complexity is entirely abstracted away. They log into the Rafay Self Service Portal and click "Deploy" for the Speech-to-Text Virtual Machine.

In a few minutes, the system automatically:

• Boots a Confidential VM.
• Triggers the Fortanix attestation client.
• Validates the security posture via the Fortanix CCM.
• Securely injects the necessary keys from the Fortanix DSM.

The result is a running NVIDIA Parakeet model in the confidential VM, protected within a hardware-encrypted enclave, ready to process sensitive financial data.

Why This Matters

This integration represents more than just a technical milestone; it is a blueprint for the Sovereign AI Factory. By removing the friction between "high security" and "high velocity," Fortanix and Rafay allow enterprises to:

Innovate with Confidence

Deploy the latest AI models (like those from NVIDIA or other proprietary vendors) without risking data exposure.

Meet Regulatory Demands

Automatically generate the attestation reports and audit trails required by compliance officers.

Scale Operationally

Move from a single pilot project to a global AI deployment using repeatable, scalable processes.

As AI continues to permeate every facet of business, the ability to run these workloads in a Confidential, Secure, and Sovereign manner will be the defining characteristic of successful digital transformation. The Fortanix and Rafay integration is leading that charge.

‍