How AI on CoreWeave is transforming the way Aston Martin Aramco competes in F1

AI is no longer theoretical. It is being utilised every day by the team at Aston Martin Aramco to challenge thinking and uncover new insights. Aston Martin Aramco Commercial Technology Ambassador Eric Ernst explores how CoreWeave is working with the team to drive innovation, a topic he will discuss on stage at Fully Connected, CoreWeave's user conference taking place in San Francisco 29 September to 1 October.

F1 has always been a sport of marginal gains, meaning the teams capable of extracting the most value from every piece of information have an increasingly powerful advantage.

Today, that information extends far beyond what can be measured in a wind tunnel or captured by a stopwatch. Millions of data points are generated every race weekend.

The challenge is no longer simply having access to that information. It is knowing what to do with it – and doing so quickly enough to make a difference. That is where AI is beginning to change the F1 landscape.

At Fully Connected 2026, Aston Martin Aramco Commercial Technology Ambassador Eric Ernst will join the conversation about what enterprise AI looks like when it has to perform in the most competitive environment on the planet.

Here, he explains what attendees can expect from the event and lifts the lid on how the team is using AI on CoreWeave to gain a competitive advantage.

Fully Connected

"Fully Connected will pull back the curtain on what enterprise AI looks like when it's deployed at the edge – in our case, trackside. It will gather leading minds in the industry: Chief Technical Officers, engineering leaders, and data scientists, all of whom are tired of hearing about theoretical AI and want to see how an elite organisation like an F1 racing team deploys it under extreme pressure.

"For Aston Martin Aramco, that pressure provides a uniquely demanding testbed. The team is working with CoreWeave to turn vast quantities of complex, unstructured information into actionable insights that can influence decisions – sometimes days before a race, sometimes during the race in the seconds available to an engineer on the pitwall.

"F1 is no longer just about who has the best wind tunnel; it's about who can extract the most value from all available information. CoreWeave provides both the infrastructure and the software that enables us to push boundaries."

At Fully Connected, you can expect insight into an array of real-world examples of AI's benefit to enterprise.

But, to whet the appetite, Eric explains just two instances where AI on CoreWeave is already helping our team.

1 | Closing the gap between prediction and reality

"A huge challenge and considered a holy grail in modern F1 aerodynamics is correlation," explains Eric.

"The CoreWeave Air Tunnel provides incredibly precise measurements, but it simply cannot replicate elements like the varying kerb heights across 24 different race weekends, the unpredictable impact of weather, or how the car will behave when caught in the dirty air of a competitor. This creates a performance delta which is a noticeable gap between what the simulations predict the car will do and what the driver actually feels on the track.

"Aston Martin Aramco partnered with CoreWeave to build a solution to close that gap. Together, we developed analytical methods to better understand performance variations between controlled testing and track conditions. By analysing historical race data patterns alongside vehicle dynamics research, we can identify key factors that influence real-world performance.

"As a result, the team is now able to anticipate how aerodynamic components will perform on the track with much higher accuracy. In the week leading up to a race, the Vehicle Engineering team uses these insights to calibrate the car's setup. The goal is simple: when the driver heads out for that very first practice lap, the car feels and performs exactly the way we expect it to. We are closing the gap between prediction and reality."

2 | Turning the noise of F1 radio into an information advantage

"During a race, one of the challenges the engineers face is that there are 22 competitor radio channels and 15+ internal intercom channels broadcasting simultaneously," says Eric.

"If we wanted to know 'how are other drivers finding the Medium tyre?' before making a split-second pit decision, an engineer literally had to manually listen to audio or trawl through noisy transcripts looking for keywords. It was an extremely manual process that could take valuable minutes.

"As our partner, CoreWeave built a tool that listens to, transcribes, and categorises audio feeds across both competitor and internal channels. We now have an interface with powerful search functions and a chat interface for broader analysis. The transcription model handles the incredibly noisy driver radios, and an LLM (Large Language Model) aligns formatting (e.g., "turn 1" to "T1") so our teams can scan transcripts much faster.

"What used to take a dedicated engineer minutes of frantic searching can now be queried and filtered instantly. We've turned unstructured, chaotic audio into structured, actionable data that helps the team make faster decisions on the pitwall.

"The scale becomes clear when looking at a single race weekend. At the Monaco Grand Prix, the system processed tens of hours of radio, thousands of messages and hundreds of thousands of words."

Turning information into action

AI isn't replacing the expertise of the people making decisions at Aston Martin Aramco. It is changing the amount of information they can process, and the speed at which they can turn it into something useful. Which brings us back to the purpose of Fully Connected.

The summit is built around precisely this shift: moving beyond conversations about what AI could do, and examining what happens when it is deployed to solve real problems at scale. For F1, those problems happen to come with a stopwatch attached.

CoreWeave

Fully Connected

To learn more, be in the room at Fully Connected, taking place at Moscone South in San Francisco from 29 September to 1 October. Tap to register your attendance.