Key takeaways from the inaugural AMR Technology Forum and the future for AI in racing.
As artificial intelligence (AI) increasingly permeates our lives and industries, it is no surprise that it has become a useful tool for a high-stakes data game like Formula 1. At every race weekend and in between, the sport generates an astonishing amount of information that requires synthesising, analysing and ultimately communicating with strategists, drivers and more.
Every lap yields approximately 60 MB of data per car, combined with an additional 120 MB processed in the garage. For the Aston Martin Aramco Formula One Team (Aston Martin F1 Team), this translates to around 1.5 TB of data exchanged with its Mission Control every week. With such a wealth of information, the challenge lies in processing it effectively.
AI is not new to the sport, but as it evolves, it presents new and innovative ways to streamline and improve the sport. On Friday 3rd of July, Aston Martin F1 Team held the inaugural AMR Technology Forum at their Technology Campus in Silverstone. Surrounded by the buzzing atmosphere of a British Grand Prix race weekend, the event marked a pivotal moment for Aston Martin, bringing together partners focused on cutting-edge technology, data, cloud solutions, and human performance to discuss the potential for AI in the sport on-and off-track.
Harnessing the power of AI
During a race weekend, engineers reference up to four years of historical data and simulations to optimise the car’s configuration. With 40 to 50 setup parameters in play and thousands of adjustable control system parameters, the teams are prepared for anything. For races like Monaco, up to 100,000 race simulations are conducted to evaluate various outcomes and optimise strategy, relying on as many as 300 variables, including driver performance and tyre conditions. In an even more remarkable statistic, data is transmitted from the track to Mission Control in 0.2 to 0.3 seconds, enabling real-time analysis.
An average Formula 1 engine control unit (ECU) completes a staggering 43 trillion calculations per race. Through supportive data infrastructure from technology partners, teams can enhance telemetry, simulations, and high-performance computing workflows.
There is an opportunity for Gen AI to create strategies and scalable cognativity power.
In a recent interview, Adrian Newey said: “We’re feeding in our own data–wind tunnel, CFD, track–and using AI to spot patterns, correlations and trends that a human might not see quickly enough. It helps us make better decisions about how to develop the car.”
Speaking with Eric Ernst at the AMR Technology Forum
The Innovation Platform Editor Maddie Hall sat down with Technology Ambassador Eric Ernst at the event.
How is AI used during races? What specific types of data do these algorithms analyse in real time and how do they influence decision-making?
You can split AI into two parts – the generative AI that can create strategies or decision-making trees. We have different data points that come in, and the engineers can play triage and select the appropriate strategy. Then there’s the machine learning aspect, where you have different models that you can train on different things. For example, non-destructive braking, or if you look at driving skills or tyre performance, using that model to identify if certain patterns are going to come up and if you can predict them. Just by using fuel, the car gets lighter every lap by 1.5 litres, so the entire driver dynamic of the car changes constantly. There are models that are created way before the race through the simulator, with the driver training for different scenarios to determine how we are going to set the car up.
Beyond the racetrack, how does Aston Martin use AI in its operations?
Beyond the racetrack, it’s like any other company. You have AI within the business operations doing what AI does in any office. Whether that be summarising emails, transcribing meetings, or translations, you can find AI as a tool throughout the entire building and across every aspect of the team.
Have there been any challenges in implementing AI into the sport and wider business?
It’s like any new technology; it takes time to adapt and time to educate people on how to use it. There need to be safeguards in place to ensure that it’s being used in the right way.
How do you envision the role of AI and other advanced technologies shaping the landscape of F1?
It really is a case of ‘the world’s your oyster.’ What’s important is that it will always stay a human-driven sport, and AI is not going to replace human involvement. The moment we take the pilot away, it won’t be a sport anymore. We won’t be working less, but rather able to do a lot more. People will be included in moderation and take on more of a triage role. Regulations will likely be introduced to define how far you can go with AI in the sport, but there’ll be a great deal that can be advanced to make the sport more sustainable and exciting to watch. That could be anything from using AI to deliver the product to the viewer, to how something is done on track, to the logistics of how equipment is packed for transport.
Conclusion
The continued integration of AI in Formula 1 and beyond presents countless opportunities, combining vast amounts of data with advanced technology to drive performance and strategic decision-making. From real-time analytics during races to enhancing operational efficiency off the track, AI is becoming an indispensable tool. As Eric Ernst emphasised, it’s about “building an AI ecosystem” and “not just adapting to the future but architecting it.”
Please note, this article will also appear in the 27th edition of our quarterly publication.
