AMD’s 4.4 GHz chips drive Kodiak’s driverless trucks to scale
AMD and Kodiak announced the partnership on Wednesday, confirming that the chipmaker’s server-grade EPYC processors will power the autonomous trucking firm’s latest hardware. The move extends AMD’s push into physical AI workloads, where real-time sensor data from cameras, lidars, and radars must be ingested, cleaned, and preprocessed before the vehicle’s software can act.
Kodiak’s SensorPods generate a continuous stream of data every second. The EPYC processors aggregate and preprocess this data, ensuring the Kodiak Driver, its autonomous software, receives low-latency inputs for tasks like sensor processing, localisation, and path planning. The choice of EPYC was driven by its I/O capabilities and clock speed, rather than raw core count. With 80 PCIe lanes and a base frequency of 3.15 GHz (boosting to 4.4 GHz), the processors deliver a 25% clock-speed gain over the previous generation while consuming less power.
Autonomous trucks operate in latency-sensitive environments where sensor data must be processed in real time. AMD’s EPYC processors are designed for precisely this kind of workload, offering the performance needed to handle the enormous volumes of data generated by autonomous vehicles.
Don Burnette, founder and CEO of Kodiak, emphasised the importance of this capability.
Kodiak’s deployment of EPYC processors comes as the company scales its operations. In the second quarter of Q2 2026, Kodiak deployed seven additional driverless trucks, bringing its total to 35 customer-owned vehicles by the end of the quarter. The company has logged more than 40,000 cumulative hours of paid driverless operation, primarily in the Permian Basin, where its trucks operate without a human in the cab. The compute upgrade is a critical step ahead of its planned expansion to public highways later this year.
Commercialising autonomous trucking with off-the-shelf hardware: Kodiak
Kodiak’s decision to use commercially available hardware from AMD reflects its focus on scalability and production-ready solutions. By leveraging off-the-shelf silicon, Kodiak can allocate more engineering resources to refining its autonomous software and expanding its fleet. This approach is particularly important as the company grows its driverless operations in the Permian Basin and prepares for public highway deployments.
Kodiak Driver is already operational in long-haul trucking, industrial trucking, and defence sectors. Commercial partners include Atlas Energy Solutions, IKEA, J.B. Hunt, Bridgestone, Werner Enterprises, C.R. England, General Dynamics Land Systems, and Roehl Transport.
Kodiak is demonstrating what’s possible when advanced autonomous driving software is paired with high-performance AMD compute. Together with Kodiak, AMD is providing the processing capabilities needed to move driverless trucking from innovation to significant scale.”
Sign up free to ask CAMAL AI for a summary, the key points or anything else about this story.
Related stories
CAMAL AI
Financial Ports Newsletter
The maritime economy, every morning
Ports, shipping and freight markets in one short email. Free.
