Future
Artificial intelligence in motorsport: Could AI make F1 faster and fairer?
by
6min read
Artificial Intelligence (AI) is a powerful tool that has its uses across numerous industries - and it could boost laptimes, safety and parity in motorsport such as Formula 1.
Sign up for a weekly newsletter and we'll make sure you're fully up-to-date in the world of race technology
The term used is ‘AI’ but a more accurate description for its use in motorsport might be ‘machine learning’. The terms are used interchangeably but there is a distinction:
- AI is a wider concept that enables technology to allow a machine to sense and reason like a human being.
- Machine learning is a part of AI that uses algorithms to analyse large data subsets, learn from that data, and make decisions.
Machine learning algorithms need ‘training’, just like a human brain, and so they improve over time. Right now, machine learning is in its infancy. You might have noticed mistakes in AI-generated images and even Google search responses: their answers to your queries are an amalgamation of thousands of images and web pages.
There’s an understandable air of caution around the technology but, within motorsport, machine learning could be the key to making fairer steward decisions.
The safety car turning the track at Sao Paulo during the 2024 Brazilian Grand Prix
How could AI be used in F1 stewarding?
Motorsport stewarding is a difficult job. With numerous cars circling the track at high speed, it’s easy to miss an incident.
Currently, the FIA has race officials at each of its races: stewards and a race director. They will sit in a room with live video screens and CCTV footage of the circuit and during an event they’ll have multiple decisions to make.
Stewards must decide if red or yellow flags need to be flown, whether a competitor must be blue-flagged, or if a safety car needs to be deployed.
They must also look at track limits violations to determine whether a driver has strayed beyond the lines too many times or cut corners, and decide if any driver is to blame for a collision. Potential pitlane infringements only add to the challenge.
With thousands of data sets available in the form of video, telemetry and images, machine learning could be used to help stewards sift through the information. FIA single-seater operations and junior categories director Francois Sicard explained to Raceteq how AI could be used in F1, Formula 2 and Formula 3.
“It can help race control for example to spot any track limits situation. Artificial Intelligence can help you detect [transgressions]. At the moment it’s done by human beings, but in the future, you can imagine that it’s done by Artificial Intelligence.”
At the moment we are only at the start, seeing what we could do and what we need to do.
Francois Sicard
, FIA single-seater operations director
Inside the race control room during the 2022 Brazilian Grand Prix
Sicard added that AI would not replace its decision-makers, but help them make the fairest decisions at a faster rate.
“It could help guide them. It’s not going to provide them the answer but it could give them guidance towards the decision they have to take.”
However, AI stewarding is far from being implemented in F1 - and the FIA refrained from offering a potential date for its debut.
“We are not using it yet but we are starting to experiment, and this is definitely something that we are starting to develop, but AI can be a good tool for the future to help race control and stewarding for F1, F2 and F3. So, we are really at the start of the process but this is really something we are starting to put some resources on,” clarified Sicard.
“At the moment we are only at the start, seeing what we could do and what we need to do. We are exploring it.”
The stricken cars of Ollie Bearman (L) and Pepe Marti (R) after a collision in the 2024 Belgian Formula 2 feature race. AI stewarding could be used in junior single-seater stewarding too
How is AI currently being used in motorsport?
One of the first motorsport teams to publicly admit to using AI is United Autosports, a sportscar racing team based in the United States.
From 2022, it utilised Valkyrie AI, a machine-learning model, to streamline its decision-making processes for determining which parts should be used and replaced on its cars.
Furthermore, the team also used the technology to analyse timing and scoring patterns from endurance races and predict when safety cars and yellow flags were most likely to occur.
McLaren F1 Team also spoke about the use of AI in late 2024 and its various applications.
McLaren race engineer Tom Stallard sitting in front of screens at the McLaren pitwall during the 2024 Saudi Arabian Grand Prix
Its lead data scientist Anjum Sayed told the team’s official website that simulations are run using machine learning to find out how each set-up configuration and combination of parts will affect the car balance and give drivers Lando Norris and Oscar Piastri a better set-up baseline.
“Tying all of these together requires thousands and thousands of simulations, most of which won’t be anywhere near the sweet spot we’re trying to locate - but using Dell Technologies’ AI Factory to join the dots helps us narrow down the possibilities and puts us in the best position to give Lando and Oscar a car that’s in the right area when practice begins,” he said.
Mercedes’s technical director James Allison said in mid-2024 that every team has begun to use AI in F1.
“The areas where it's already kicking around the place are in aerodynamics, a bit in race strategy and it's increasingly being used to create surrogate models, to create a filter of ideas, so that you can hopefully run your ideas through a trained machine learning cluster that will give you an idea whether it's better than your last idea or give you a higher percentage chance that that idea might be slightly better than the baseline.
AI is not there to replace anybody. AI is there to add to our data scientists and engineers and helps inform their decisions.
Edward Green
, McLaren F1 Team head of commercial technology
Lewis Hamilton pits his Mercedes in the 2024 Hungarian Grand Prix. Mercedes’s technical director James Allison says every F1 team is utilising AI to some degree
“And therefore, if you make that work effectively, then gradually your rate of finding laptime gain will be somewhat steeper. So, it's already in all the teams, and it will gather pace in the seasons to come.”
McLaren also said that it uses AI to help its engineers determine the optimum strategy for a race. However, the F1 cost cap (set at $135 million USD for 2025) limits how much F1 teams can explore and use the technology.
It’s clear that AI and machine learning tools have their uses in F1 but, as stressed by the FIA, AI won’t be replacing engineers and strategists.
McLaren’s head of commercial technology, Edward Green, concurred: “AI is not there to replace anybody. AI is there to add to our data scientists and engineers and helps inform their decisions.”
AI and machine learning have the potential to revolutionise motorsport but there is a degree of caution being used - and rightly so - in anticipating AI’s accuracy and benefits.
As the technology matures and finds a greater role in the motorsport ecosystem, it could make the sport fairer while allowing teams to make decisions at a more effective rate.
When machine learning models mature and become more accurate, this will however lead to the inevitable question: will the use of AI and machine learning eventually need to be policed in the FIA’s technical and sporting regulations?
Main image by JustFormulaCar