Deepgram continues to roll out extensions to its real-time voice AI platform.
- Deepgram and Fortanix are the first to bring confidential computing to real-time voice AI, allowing organizations to protect sensitive conversations and AI models even while they are actively being used.
- Until now, you could protect data when it was stored and when it was moving. The hard part was protecting it while the AI was actually using it. That’s what we’re solving here for real-time voice AI.
- Sensitive enterprise data and proprietary model IP remain private during active inference, with no exposure to underlying infrastructure
From the press release:
Deepgram and Fortanix announced a partnership that will enable enterprises to run voice AI in their own environment on their own terms while ensuring their most sensitive data is securely protected. Under terms of the agreement, Deepgram can leverage Fortanix Confidential AI and NVIDIA Confidential Computing to add an additional layer of advanced security to self-hosted environments to ensure that its proprietary model weights, built on business-critical intellectual property, can be deployed while protecting against model theft or inappropriate use.
For enterprises, especially those in highly regulated industries, security requirements continue to tighten. Organizations handling patient conversations, financial transactions, or classified information increasingly require that sensitive audio and AI model weights remain protected not only at rest and in transit, but also during active processing in their own environments. This level of protection enables organizations to build highly-secure real-time voice applications without sacrificing on performance.
The on-premises solution runs Deepgram’s voice AI models with Fortanix Confidential AI on NVIDIA Confidential Computing-enabled GPUs, creating a hardware-isolated environment where both audio data and model weights remain encrypted and protected throughout active use. NVIDIA GPUs with Confidential Computing enable AI workloads to process sensitive data inside a trusted execution environment — a capability traditional infrastructure cannot provide. By bringing together best-in-class voice AI models, hardware-rooted isolation, and a jointly engineered, pre-integrated stack, the partnership delivers a level of in-use data protection that, until now, has not been practical to deploy at enterprise scale.
