NVIDIA is ushering in a new era of AI-defined driving, bringing its advanced NVIDIA DRIVE AV software to U.S. roads by the end of this year. The rollout begins with Mercedes-Benz, a long-time partner in the pursuit of safer, smarter mobility.
At the center of this milestone is the all-new Mercedes-Benz CLA — the brand’s first vehicle built on the MB.OS platform. The CLA introduces next-generation driver-assistance features powered by NVIDIA’s full-stack DRIVE AV software, AI infrastructure, and accelerated computing technology.
This software-defined approach allows Mercedes-Benz to deliver over-the-air updates, unlocking future upgrades and new features without requiring hardware changes. Planned enhancements to MB.DRIVE ASSIST PRO may be available both ex-factory and through the Mercedes-Benz Store, ensuring the vehicle continues to improve over time.
The CLA’s safety credentials are already impressive. It recently earned a five-star EuroNCAP safety rating, with MB.DRIVE active safety features playing a key role in accident avoidance and mitigation — a major contributor to its top safety score.
“As the automotive industry embraces physical AI, NVIDIA is the intelligence backbone that makes every vehicle programmable, updatable, and perpetually improving through data and software,” said Ali Kani, Vice President of Automotive at NVIDIA. “Starting with Mercedes-Benz and its incredible new CLA, we’re celebrating a stunning achievement in safety, design, engineering, and AI-powered driving that turns every car into a living, learning machine.”
NVIDIA DRIVE AV: Dual-Stack Intelligence for Safety and Performance
At the core of NVIDIA DRIVE AV is a dual-stack architecture designed to balance innovation with safety. An end-to-end AI driving stack handles core driving functions, while a parallel, classical safety stack — built on the NVIDIA Halos safety system — provides redundancy and robust safety guardrails.
This design enables vehicles to learn from vast amounts of real-world and synthetic driving data, helping them navigate complex environments with human-like decision-making. For drivers, this translates into greater confidence and comfort, knowing the system includes fail-safe checks and operates within clearly defined safety parameters.
The unified architecture supports advanced Level 2 driver-assistance features, including point-to-point urban navigation, proactive collision avoidance, automated parking in tight spaces, and cooperative steering that seamlessly blends driver input with system control.
Human-Like Urban Driving Powered by AI
NVIDIA’s deep learning models are driving a new generation of intelligent urban driving systems. These AI models interpret traffic holistically, allowing vehicles to:
– Navigate complex city environments with intelligent lane selection, turns, and route-following
– Detect and respond proactively to vulnerable road users such as pedestrians, cyclists, and scooter riders
– Assist drivers with point-to-point navigation, enabling safe trips from any address to any destination
The result is smoother, more intuitive assistance that adapts to real-world driving conditions.
Accelerating Manufacturing With Digital Twins
Beyond the vehicle itself, NVIDIA and Mercedes-Benz are transforming automotive manufacturing through a digital-first approach powered by NVIDIA Omniverse.
By using digital twins of factories and assembly lines, engineers can design, test, and optimize production processes virtually. This reduces downtime, accelerates iteration, and improves efficiency. Omniverse, together with the NVIDIA Cosmos platform, also enables developers to validate intelligent driving software in simulated environments before deploying it on real roads.
Cloud-to-Car Development: NVIDIA’s Three-Computer Architecture
All NVIDIA-powered intelligent driving systems rely on a cloud-to-car development pipeline that converts real-world driving data into billions of simulated miles:
– Training Infrastructure: NVIDIA DGX systems use large-scale GPU computing to train DRIVE AV foundation models on diverse global datasets, capturing human driving behavior across millions of scenarios.
– Simulation and Validation: NVIDIA Omniverse and Cosmos enable physically accurate simulations, allowing developers to test thousands of edge cases before deployment.
– In-Vehicle Compute: NVIDIA DRIVE AGX and the DRIVE Hyperion architecture handle real-time perception, sensor fusion, and decision-making while providing sensor redundancy for enhanced safety.
This closed-loop approach accelerates development, improves accuracy, validates safety in rare or hazardous scenarios, and supports scalable deployment across multiple vehicle platforms.
The Era of AI-Defined Transportation Has Arrived
Mercedes-Benz is just the beginning. NVIDIA collaborates with automakers worldwide to deliver full-stack software, AI infrastructure, and advanced computing across a wide range of vehicles.
By simplifying development and enabling continuous upgrades, NVIDIA’s approach allows manufacturers to bring intelligent driving capabilities and safety innovations to market at scale — marking a decisive step into the era of AI-defined transportation.
