Fortanix Confidential AI Protects Proprietary Model IP and Data for Secure AI Inference in Enterprise AI Factories

Mutual trust unlocks real AI outcomes using highly sensitive data and proprietary AI models without exposing assets to infrastructure operators, cloud providers or unauthorized access

SANTA CLARA, Calif.--(BUSINESS WIRE)--Fortanix® Inc., global leader in data and AI security and a pioneer of Confidential Computing, today announced a new Confidential AI solution powered by NVIDIA Confidential Computing that enables model developers to securely distribute models for deployment in on-premises AI factories without the risk of model theft or inappropriate use. The solution also enables organizations to securely run third-party proprietary AI models within their infrastructure while ensuring their data stays local and in their control.

With Fortanix Confidential AI and NVIDIA Confidential Computing, enterprises can now run best-in-class AI on their most sensitive data without exposing that data to the model vendor, and without the model vendor exposing the enterprise's intellectual property. What was previously blocked by mutual considerations of protecting IP and data privacy is now enabled by mutual verifiable cryptographic proof.

Running Proprietary Models on Sensitive Data Without Exposing Either

Enterprises can now run AI inference at scale and balance model security and data privacy. Fortanix Confidential AI creates a trusted execution environment where proprietary model weights remain encrypted and invisible, even to the infrastructure running them, while sensitive input data and the model itself are never exposed.

This means model owners can deploy their IP into enterprise environments without fear of extraction or replication, and enterprises can run inference on their most regulated data without violating privacy obligations. Both parties operate with cryptographic guarantees, not contractual ones. The result: AI adoption at scale with trust, security, and sovereignty at the core.

When enterprises deploy third-party proprietary AI models on confidential computing chips in their environment, both infrastructure operators and model providers attain strong assurances that:

• Cryptographic keys are released only to verified, trusted runtimes
• Model weights remain confidential and tamper-proof
• Customer prompts and outputs are encrypted in memory
• Deployment environments can’t be altered without detection

Fortanix Confidential AI addresses these challenges by combining Confidential Computing, secure key release and centralized policy enforcement to secure AI workloads during runtime.

“AI security can break during inference if you don’t protect data and models in use,” said Anuj Jaiswal, Chief Product and Strategy Officer, Fortanix. “This deployment on NVIDIA Confidential Computing-backed GPUs, verified by Fortanix Confidential Computing Manager and backed by secure key release from Fortanix Data Security Manager, demonstrates that you no longer have to choose between performance and protection. Confidential AI enables both.”

Organizations can now select best-of-breed third-party AI models to drive meaningful experiences and outcomes, while AI model owners can safely monetize and expand their business by sharing their proprietary models for deployment on premises with full confidence that they're not exposing their IP.

How Confidential AI Deployment Works

Fortanix Confidential AI is the industry's premier protector of proprietary models, data and inference across the AI lifecycle. It enforces:

• Hardware-based validation of the runtime
• Secure key release tied to verified workloads
• Centralized cryptographic policy control

Ultimately, the model provider’s IP is protected, and enterprises can confidently provide AI services without handling (and potentially exposing) plaintext model assets, and end users’ prompts and generated outputs remain secure.

“The next phase of enterprise AI adoption requires a foundation of verifiable trust to ensure both data privacy and proprietary model integrity,” said Anne Hecht, Senior Director AI Platforms at NVIDIA. “The integration of NVIDIA Confidential Computing and Fortanix Confidential AI enables customers to deploy AI with security and privacy.”

"Our models represent years of proprietary research and engineering - protecting that IP while expanding access is a core tension in enterprise deployment,” said Kuba Abramczyk, Forward Deployed Engineer at ElevenLabs. “Working with Fortanix on NVIDIA Confidential Computing-backed infrastructure lets us resolve that directly, giving organizations in government, healthcare, and finance the ability to run our Text to Speech models on their own servers, on their own data."

Enabling Secure AI Factories

The collaboration highlights a broader shift as AI becomes a cornerstone of enterprise operations: Confidential computing is a prerequisite for scalable AI deployment. By securing model weights, cryptographic keys and in-memory data, Fortanix enables:

• Secure AI model marketplaces
• Multi-party AI collaborations
• Regulated industry AI deployments
• Protection of proprietary foundation models
• Quantum-resilient cryptographic enforcement

Available immediately, Fortanix Confidential AI ensures that both model providers and enterprises maintain control and trust in increasingly distributed AI ecosystems. Organizations seeking to securely deploy generative AI models in external or hybrid environments can learn more at https://www.fortanix.com/platform/confidential-ai.

About Fortanix

Fortanix is the global leader in data and AI security and a pioneer of Confidential Computing, delivering a unified platform to protect sensitive data, AI models, and applications across on-premises and multi-cloud environments—at rest, in transit, and in use. Built on hardware-enforced security, Fortanix enables workloads to run in tamper-proof, isolated enclaves, protecting against data leakage, model extraction, and unauthorized access, even from privileged insiders.

As enterprises scale modern AI factories, Fortanix provides a Confidential AI foundation built on attestation-based trust, secure key release, and zero-trust enforcement. This ensures AI systems remain verifiable, auditable, and trustworthy—while enabling crypto agility and long-term resilience through post-quantum cryptography (PQC).

Trusted by leading enterprises and sovereign entities in highly regulated industries and powered by an industry-standard approach to Confidential Computing supported by leading technology providers, including Intel and Microsoft, Fortanix enables organizations to safely use sensitive data and drive AI adoption with confidence.