Sequitur Labs Releases Turnkey Solution to Simplify Protection of Edge AI Models Powered by the NVIDIA Jetson Platform

By AIT News Desk On Mar 22, 2022

Latest release of EmSPARK Security Suite for the NVIDIA Jetson platform features new trusted applications for AI protection

Sequitur Labs announced a new release of its EmSPARK Security Suite for the NVIDIA Jetson edge AI platform, featuring a generally available development kit and pre-built Trusted Applications that provide robust security features and functions needed to protect AI models at the edge.

Latest Aithority Insights : Agora Data Files Patent Application for Industry-Leading Innovations

“Building a release of EmSPARK for the NVIDIA Jetson platform is part of our commitment to providing a complete array of tools for protecting AI models on edge devices”

Expanding Sequitur’s support for the NVIDIA Jetson platform, the latest release features new Trusted Applications supporting tools for protection of AI models on edge devices, and a turnkey evaluation image that can be implemented on the Jetson platform for development and integration. This new release works in concert with the NVIDIA JetPack SDK, including support for the latest version (JetPack 4.6).

NVIDIA Jetson is the world’s leading platform for autonomous machines, robotics and other embedded applications with over 850,000 developers. With pretrained AI models, developer SDKs and support for cloud-native technologies across the full Jetson lineup, manufacturers of intelligent machines and AI developers can build and deploy high-quality, software-defined features on embedded and edge devices. Cloud-native support helps manufacturers and developers implement frequent improvements, improve accuracy, and use the latest features with Jetson-based edge AI devices.

NVIDIA JetPack comes with NVIDIA container runtime and a full Linux software development environment. This allows developers to package their applications for Jetson with all its dependencies into a single container that is designed to work in any deployment. And it is powered by the same NVIDIA CUDA-X accelerated computing stack used to create breakthrough AI products in such fields as robotics, industrial IoT, healthcare, smart cities and more.

JetPack provides the tools necessary to get started in developing features and functions of edge applications. However, manufacturers still need to develop their own production-quality Trusted Execution Environment (TEE) and applications for cryptography, key management, storage, cloud integration and AI model protection. Sequitur provides all of these, pre-built for NVIDIA Jetson, with the EmSPARK Security Suite and the EmPOWER Lifecycle Management platform.

Browse The Complete News About Aithority: Bandwidth Announces Integration with Pindrop to Add Out-of-the-Box Voice Authentication for Enterprise Contact Center Move to the Cloud

Sequitur Labs’ EmSPARK Security Suite is designed to address solutions in industries where embedded security is paramount, in particular, protection of AI models at the edge. Supporting security functions for encryption, storage, data transmission and key/certificate management are delivered by EmSPARK and housed in a microprocessor’s secure memory partition. IoT hardware manufacturers use EmSPARK to easily implement device-level security by addressing technical, IP, supply chain and business process challenges. Developers can build applications that use secure resources without having to become experts in cryptography and complex chip-level security technologies.

“Building a release of EmSPARK for the NVIDIA Jetson platform is part of our commitment to providing a complete array of tools for protecting AI models on edge devices,” said Philip Attfield, co-founder and CEO of Sequitur Labs. “With this release, you can access the latest test software with all of the documentation and tools you need. It’s a turnkey solution and it’s easier than ever to quickly integrate tools for securing devices and protecting AI models on the Jetson platform.”

[To share your insights with us, please write to sghosh@martechseries.com]