Lattice Joins NVIDIA Halos Ecosystem to Pioneer Safety Standards for Physical AI

In a significant development announced at NVIDIA GTC 2026, Lattice Semiconductor has joined the NVIDIA Halos ecosystem, marking a key step toward standardized safety in physical AI systems. The collaboration aims to build Halos-certified designs using Lattice's Holoscan Sensor Bridge, focusing on applications in robotics, industrial automation, and autonomous vehicles. This move addresses the growing need for reliable, low-latency AI deployment in real-world environments where safety is paramount.

NVIDIA Halos represents a comprehensive full-stack safety system for physical AI, integrating hardware, software, tools, models, and design principles. It unifies safety elements across vehicle and robotics architectures, including their underlying AI models. Notably, the NVIDIA Halos AI Systems Inspection Lab is the first ANSI National Accreditation Board (ANAB) accredited facility for inspecting AI-driven physical systems, providing a rigorous certification framework.

Lattice contributes its expertise in low-power FPGAs and solution stacks to enable scalable, trusted physical AI. The company's hardware-accelerated Holoscan Sensor Bridge technology supports sensor-over-Ethernet processing with low latency, determinism, and efficiency at the edge. By partnering with NVIDIA and other ecosystem members, Lattice will help develop best practices for safety-certified designs.

The announcement underscores the rapid transition of physical AI from labs to deployment, where failures could have severe consequences. NVIDIA Halos promotes a unified approach to safety, reliability, and trust, aligning hardware and software for end-to-end autonomous stacks. Lattice's involvement accelerates the creation of certified systems essential for industries relying on AI-driven automation.

This partnership signals a maturing ecosystem for AI safety, particularly in physical domains beyond traditional digital alignment concerns. As more companies engage, Halos could set global standards, mitigating risks in high-stakes applications and fostering broader adoption of safe physical AI technologies.