“The autonomous vehicle revolution has arrived, and automotive will be one of the largest AI and robotics industries.”
Jensen Huang, Founder and CEO, Nvidia
The Consumer Electronics Show (CES) 2025 in Las Vegas has set a new benchmark for technological innovation, particularly for autonomous vehicles (AVs). This year’s event was not just a showcase of cutting-edge gadgets; it was a clear signal of the seismic shift anticipated in the global economy, with Nvidia Corp.’s CEO, Jensen Huang, boldly proclaiming that “autonomous vehicles will likely be the first multi-trillion dollar robotics industry.”
Huang’s keynote at CES 2025 was filled with both visionary insights and concrete statistics that underscored this prediction. He highlighted that there are currently over a billion vehicles on the road globally, collectively driving more than a trillion miles each year. With the advent of AI-driven autonomy, Huang forecasted that “in the future, many of these vehicles will be heavily autonomous,” potentially transforming the automotive sector into a colossal economic force.
During the keynote, Huang also introduced several groundbreaking technologies aimed at accelerating the development of autonomous systems, including Nvidia Cosmos, a platform designed to simulate physics-aware environments for training robots and AVs. This platform, alongside Nvidia’s DRIVE Hyperion system, supports the industry’s move towards creating vehicles that not only drive themselves but do so with unprecedented safety and efficiency.
Huang’s assertion was further backed by Nvidia’s reported revenue surge, reaching over $35 billion in the company’s most recent fiscal quarter, up from the $27 billion for the entirety of fiscal 2023, showcasing the growing market demand for AI and robotics solutions.
Also, Nvidia opined that it expects its automotive business to expand to $6bn in the 2026 financial year.
What It Takes to Build an Autonomous Vehicle
According to Jensen Huang, Nvidia’s CEO, building autonomous vehicles, like all robots, requires three computers: NVIDIA DGX to train AI models, Omniverse to test drive and generate synthetic data, and DRIVE AGX, a supercomputer in the car,” Huang explained, detailing the technological ecosystem Nvidia is fostering.
Nvidia’s Drive Thor Processor
Nvidia’s Thor processor, formally known as the “NVIDIA DRIVE Thor” system-on-a-chip (SoC), is a significant advancement in the realm of automotive AI computing. Unveiled at CES 2025, Thor is designed to be the brain behind next-generation autonomous vehicles (AVs), offering unprecedented computational power for real-time processing of sensor data, which is critical for autonomous driving.
“The Thor chip, with 20 times the processing power of its predecessor, integrates data from sensors like cameras, radars, and Lidars to enable real-time autonomous navigation and serves as a universal robotics computer for various applications,” Huang noted during his keynote address.
Power and Specifications of Nvidia’s Thor Processor
Computational Power: Thor is reported to have 20 times the processing power of its predecessor, the NVIDIA DRIVE Orin. This increase in capability is crucial for handling complex algorithms and vast amounts of data from multiple sensors, like cameras, radars, and LiDARs in real-time.
AI Performance: It achieves over 1,000 teraflops of performance, making it capable of running sophisticated AI models necessary for advanced autonomous driving features. This includes not only perception but also prediction, planning, and control of vehicle movements.
Architecture: Thor is built on Nvidia’s Blackwell architecture, which is known for its efficiency in AI workloads. It includes a generative AI engine, further enhancing its capabilities for creating dynamic, real-world simulations for testing and development.
Role in Autonomous Vehicles
Thor serves as a universal robotics computer, not limited to automotive applications but versatile enough to be used in various robotic systems. Here’s how it specifically enhances autonomous vehicle functionality:
Sensor Fusion and Processing: Thor integrates and processes data from a rich array of sensors to create a comprehensive view of the vehicle’s surroundings. This fusion of data allows for precise object detection, classification, and decision-making in complex driving scenarios.
Real-Time Decision Making: The high computational power enables real-time processing of this data, allowing vehicles to make split-second decisions regarding navigation, obstacle avoidance, and route planning.
Safety and Reliability: With the Thor processor, Nvidia has achieved ASIL D certification for its DRIVE OS, which is the highest standard for automotive functional safety. This certification ensures that the system is reliable enough for critical driving operations, with Huang stating, “NVIDIA Drive OS has achieved ASIL D certification, ensuring that our autonomous driving technology meets the highest safety standards.”
Simulation and Training: Thor supports NVIDIA’s Omniverse and Cosmos platforms for simulation. These tools can generate “mountains of training data for autonomous vehicles,” as Huang noted during his CES 2025 keynote. This synthetic data is vital for training AI models in various scenarios without the need for extensive real-world testing.
Partnerships and Developments
Toyota and Nvidia
At CES 2025, Toyota announced a significant expansion of its partnership with Nvidia to integrate Nvidia’s technology into its next-generation vehicles. Toyota will utilize Nvidia’s DRIVE AGX Orin supercomputer along with Nvidia’s safety-certified DriveOS operating system. These technologies aim to provide functionally safe, advanced driving assistance features, setting the stage for Toyota’s autonomous vehicle ambitions.
Nvidia’s CEO Jensen Huang stated, “Toyota is actually a great example of our cloud-to-car strategy,” highlighting the transition from cloud-based AI model training to in-car implementation.
The partnership was well-received by investors, with Toyota shares rising nearly 4% post-announcement, reflecting market confidence in Toyota’s strategy to integrate high-end AI and computing for vehicles.
Aurora Innovation and Nvidia
Aurora Innovation, in collaboration with Nvidia and automotive supplier Continental, announced a long-term strategic partnership to deploy driverless trucks at scale. Aurora will integrate Nvidia’s DRIVE Thor system-on-a-chip (SoC) and the DriveOS into the Aurora Driver, which is an SAE level 4 autonomous driving system. This integration is planned for mass production by Continental in 2027.
Chris Urmson, CEO of Aurora, emphasized the monumental shift towards autonomy, saying, “Delivering one driverless truck will be monumental. Deploying thousands will change the way we live.” underscoring the transformative potential of this partnership.
The partnership signifies a significant step towards commercializing autonomous trucking, potentially revolutionizing logistics with improved efficiency, safety, and cost savings.
Aurora Innovation’s shares surged over 35% following the announcement, indicating strong market approval and investor enthusiasm towards the partnership’s prospects.
Nvidia: A Technological Backbone
Nvidia stands out as the technological backbone, providing not only hardware like DRIVE AGX Orin and Thor but also the software (DriveOS) that enables sophisticated AI processing for autonomous driving. Nvidia’s strategy here is to be the preferred platform for automotive AI, mirroring its success in other computing markets.
The Autonomous Vehicle Industry in 2025
The autonomous vehicle industry in 2025 is characterized by significant advancements, with Alphabet’s Waymo holding a leadership position in terms of operational deployment and technological maturity. Waymo has leveraged its years of experience since its inception as Google’s self-driving car project to become the first to offer a widespread commercial robotaxi service. With over 150,000 autonomous paid rides completed on a weekly basis, Waymo has not only demonstrated the feasibility of driverless taxis but has also scaled this service across multiple US cities, most notably Phoenix, San Francisco, Los Angeles, and Austin.
While not yet offering a fully autonomous, driverless service, Tesla has made notable strides in driver assistance technology. Tesla’s Full Self-Driving (FSD) system, although still requiring active supervision, has shown improvements in handling complex driving scenarios across diverse environments. In October 2024, Tesla unveiled a prototype of its fully driverless vehicle, the “Cybercab,” at the “We, Robot” Tesla event. Elon Musk and Tesla are looking at 2026 for mass production.
Amazon’s Zoox has taken a different path by designing vehicles from the ground up specifically for autonomous ride-hailing. Unlike retrofitting existing cars, Zoox’s bidirectional, purpose-built vehicles without a steering wheel represent a bespoke solution to urban mobility challenges. While Zoox has been slower to roll out public services, it has made significant progress in testing and has secured permits to operate driverless in select areas like Foster City and Las Vegas. At CES 2025, a couple of attendees had the opportunity to take autonomous rides in the Zoox robotaxi.
Conclusion
CES 2025 was not just an event but a milestone, marking the dawn of an era where autonomous vehicles could redefine mobility, commerce, and urban landscapes. With Nvidia taking center stage, autonomous vehicles are set to lead the charge in the robotics industry, potentially reshaping our understanding of transportation and economics on a global scale.
You May Also Like:
Top 5 US Autonomous Vehicle Stocks You Should Buy in 2025
U.S. Autonomous Vehicles Roundup: November/December 2024
GM Stops Funding Its Cruise Robotaxi Project, as Waymo, Zoox Race Ahead
Women in the U.S. are Turning to Driverless Taxis Over Rideshare Services. Here’s Why!
I’m Dr. Brandial Bright, also known as the AVangelist. As a dedicated and passionate researcher in autonomous and electric vehicles (AVs and EVs), my mission is to educate and raise awareness within the automotive industry. As the Founder and Managing Partner of Fifth Level Consulting, I promote the adoption and innovation of advanced vehicle technologies through speaking engagements, consulting, and research as we progress to level 5 fully autonomous vehicles.
