Aston Martin F1 Uses AI to Monitor Rival Radio – PYMNTS.com
Highlights
The pit wall now hears every channel. CoreWeave-built AI transcribes and sorts 40 live radio channels within five seconds of each message.
Simulation is replacing some trial and error. Models trained on high-fidelity airflow simulations estimate the effect of design changes without running every full simulation.
AI is spreading from the car to the office. Cohere’s North platform is available to every team member, and Cognition’s agents handle software development tasks.
Aston Martin’s Formula One team now listens to 40 race radio channels at once, with artificial intelligence transcribing and sorting each message within five seconds, CoreWeave said. The system lets a race engineer ask how rival drivers are finding a tire compound and get an answer mid-race.
Complete the form to unlock this article and enjoy unlimited free access to all PYMNTS content — no additional logins required.
Subscribe to our daily newsletter, PYMNTS Today.
By completing this form, you agree to receive marketing communications from PYMNTS and to the sharing of your information with our sponsor, if applicable, in accordance with our Privacy Policy and Terms and Conditions.
The Aston Martin Aramco Formula One Team carries the luxury carmaker’s name and races out of Silverstone, England. It has added three AI partners in the past 18 months. CoreWeave handles AI cloud computing and race data, Cohere handles generative AI across engineering and operations and Cognition supplies AI coding agents.
Every Formula One team can legally listen to its rivals’ radio. Few can use it in time. At the 2026 Monaco Grand Prix, teams spoke more than 380,000 words across 50 hours of audio and 21,000 messages, according to CoreWeave. Before the new system, answering a question about rival tire wear meant pulling an engineer off other duties to search transcripts, a process that took several minutes.
CoreWeave engineers worked on site with the team from October 2025 to build the platform. It covers 22 competitor channels and the team’s own internal channels. The model went through 75 iterations and trained on more than 3,000 labeled samples from real race audio, CoreWeave said. It ran as a test at the final two races of 2025 and went into full use for the 2026 season. CoreWeave’s figures are self-reported.
We’d love to be your preferred source for news.
Please add us to your preferred sources list so our news, data and interviews show up in your feed. Thanks!
Formula One caps what teams can spend on car performance at $215 million for 2026, Formula1.com reported. The FIA also rations wind tunnel time. Aston Martin uses models trained on a smaller set of high-fidelity airflow simulations to estimate the effect of design changes without running every full simulation, TechInformed reported in July. The same approach speeds analysis of wind tunnel data.
Falling compute costs help the team work within the cap. “Every 12 to 18 months, the crunch gets cheaper,” Aston Martin Chief Information Officer Fabrizio Pilotti told TechInformed. The savings go back into engineering instead of infrastructure, he said.
AI is also moving onto the car. One project aims to replace a heavy optical speed sensor with a machine learning model that estimates sideslip from wheel speed, ride height and suspension data. Another reads tire and brake condition from onboard thermal cameras. Chipmaker Arm, the team’s AI compute partner, said its silicon lets the team process 10 times more aerodynamic data in real time.
Cohere joined in March as the team’s official generative AI partner. Every team member gained access to North, Cohere’s agentic AI platform, the team said in its announcement in March. Cohere will support performance, engineering and operations work. Cohere’s Ryan Lewis said the goal is to let staff offload routine tasks and focus on race decisions.
Cognition signed a multi-year deal in February. Its AI agents will perform software development tasks autonomously across the team’s existing infrastructure, according to the team’s February announcement.
Aston Martin is not alone. AI and machine learning brands account for 4 of the top 15 new sponsorship investors in the sport, according to SponsorUnited data, PYMNTS reported in June.
Blockchain.com Seeks US Approval for Event Contracts and Derivatives
Bending Spoons Deploys AI to Overhaul Acquired Software Companies
Banks Put AI Agents Through Performance Reviews
Mastercard CEO Says Stablecoins Solve Cross-Border Payment Delays
Get PYMNTS Today, AI, B2B and more.
Own the conversation. Drive demand.
source
This is a newsfeed from leading technology publications. No additional editorial review has been performed before posting.
