The age of artificial intelligence (AI) is upon us, and it is no surprise that Formula One – so often a leader in sport when it comes to technological innovation – is playing a role in its adoption.
Even as the series expands into new markets and develops new fan engagement initiatives, Formula One continues to be a testbed for technology. Even with the plethora of cutting-edge, boundary-pushing advances, road relevance still guides the global motorsport series’ decision makers.
Formula One’s move to introduce 100 per cent sustainable fuels from the start of the 2026 season, for example, was a clear response to the wider automotive industry moving towards cleaner energy solutions.
In more recent times, and just like other industries, it has been turning its attention to artificial intelligence (AI). Of course, there is no doubt that AI will become a huge part of the vehicles people drive in the future – and it is already being seen today. For example, cars can detect nearby traffic and react to potential incidents faster than the drivers, vastly increasing the safety of roads.
Formula One doesn’t use AI to only assist with safety, though. The technology is being applied far and wide within the series’ ecosystem, including in areas such as broadcasting, sustainability, and race strategies.
But how does it work in practice? BlackBook Motorsport provides an overview how AI is being used in Formula One today and how that could evolve moving forward.
A different lens
It’s impossible to talk about AI in Formula One without mentioning Amazon Web Services (AWS). A global partner of the series since 2018, the cloud computing company has played a pivotal role in the digital transformation of the series.
To give an idea of just how technical Formula One is, there are more than 300 sensors on each car generating over 1.1 million data points per second. The role of condensing this data and presenting it in a digestible way falls to AWS – and AI is the tool that simplifies the process.
For broadcasting, AI can be used to both make backend operational processes more efficient and generate stats that ensure Formula One coverage is more engaging and informative for viewers.
While not exclusive to broadcasting, an example of the former is ‘Root Cause Analysis’, which uses natural language processing to investigate system errors. The programme looks out for if a switch in a container is repeatedly failing, which the relevant technicians would previously be unaware of until having to use it.
The use of AI avoids this scenario, cutting off the problem before it becomes a wider issue. The next step is training the AI to proactively search for errors, rather than analysing trends and highlighting things that have failed in the past.
Related posts
- How AWS is supporting F1’s sustainability drive using the cloud and AI
- How AI is changing the way we watch F1
“I wouldn’t put a timescale on it [becoming proactive], but that is definitely the vision,” Pete Samara, director of innovation and digital technology at Formula One, told BlackBook Motorsport in July. “We’re pushing really hard on phase one, which is [Root Cause Analysis], and we know that once we’ve done that, we actually understand how to use it, and understand all the things we can add into the proactive mindset.”
AWS has also implemented the ‘StatBot’ tool, which uses generative AI to answer questions Formula One analysts might have about results sometimes dating back decades, such as the last time a rookie won a Grand Prix.
This simplified a manual process which would involve poring through a historic data repository, which could take up to ten minutes.
“Formula One started in 1950, now we’re in 2024,” said Neil Ralph, principal sports industry specialist at AWS. “Using StatBot against that historic data repository, you get that answer back in seconds, so it’s available for that fan engagement aspect.”
How AI is contributing to F1’s sustainable future
That digital transformation has also helped reduce the need to transport physical equipment between racetracks, supporting the series’ ongoing sustainability drive.
According to Rob Smedley, the former Williams, Jordan, and Ferrari engineer, masses of IT equipment had to be shipped from race to race only 15 years ago, but adoption of the cloud and remote production capabilities means that is no longer the case.
“We’re not having to send masses of data down a big tube all the way back to [the Formula One headquarters in] Biggin Hill,” Smedley said in August last year. “Now, we’ve got a digital archive all stored in the cloud. All the data we’re bringing off the cars and the timing systems goes into the cloud too.
“All of that has a big impact on sustainability: less kit, less people, and you’re generating less heat for that day.”
Formula One is striving to be Net Zero by 2030, and AI will therefore have a role to play in achieving this lofty goal. Even so, machine learning will likely be used to help inform the series’ sustainability strategy, especially in terms of how to speed up processes or improve their accuracy, rather than being the solution itself.
Research by PwC UK found that AI levers could reduce worldwide greenhouse gas emissions by four per cent in 2030, equivalent to 2.4 billion tonnes of CO2e emissions, illustrating how the likes of Formula One could implement the technology to improve sustainability.
In that context, this is one area where Formula One can lean on AI more, especially to maximise logistical efficiencies at a time when its global calendar is increasingly bloated. For example, AI could evaluate the most sustainable approach to a season through a mix of air, sea, and road freight.
Running the numbers
The area that will be most obvious to fans is how AI is improving the race strategies of Formula One teams.
For a number of years, teams have used Monte Carlo simulations to work out the most efficient approach to a race weekend. Pioneered during the 1990s, the process uses a plethora of historic data to work out the likely outcomes during a race, but it’s not a foolproof method.
With the ongoing development of AI, the simulations can be more complex and include a wider range information, in turn making them more accurate. For instance, it was historically quite easy for teams to run Monte Carlo simulations for their own cars, but it was very difficult to factor in the effect the presence of 18 other cars would have during the race.
It is no coincidence that Red Bull’s race strategy has been near-perfect since they agreed a partnership with tech giant Oracle, which has put the team in a position where it is able to run simulations that include their nearest rivals. It would be no surprise if the partnership is now at the stage where all 20 cars can be accurately included in the process.
“We have a lot of very data-hungry engineers,” Red Bull team principal Christian Horner told the SportsPro Podcast back in 2022, as part of a wide-ranging conversation with then-Oracle chief marketing officer Ariel Kelman. “Data is our lifeblood, how we manage that data, how we store that data. Pivotal moments are driven by data. Utilising the Oracle resource, we’re able to make a much more informed decision, whether it’s the conversion to the soft strategy in France [in 2021] that won Max [Verstappen] the Grand Prix there, or the utilisation of safety cars at the last Grand Prix.”
Indeed, these vast datasets need to be paired with an analytical mind to make key decisions. Once more, this emphasises how AI is a tool to help inform decision-making. At Red Bull, this responsibility falls to principal strategy engineer Hannah Schmitz, one of the most important contributors to the Milton Keynes-based team’s recent success.
How do the Fastest Driver rankings work?
— Formula 1 (@F1) August 20, 2020
Take a closer look at the processes involved with the latest F1 Insight powered by @awscloud#F1Insights #F1 pic.twitter.com/kD0LadZOxP
What about the future?
Four years ago, AWS named Heikki Kovalainen the eighth-fastest driver in Formula One history through the use of machine learning. Clearly talented over one lap, the Finn still only managed one win and four podiums in his 111-race career.
AI must therefore be applied with context. After all, there are very few people, if any, that would put Kovalainen in their top ten Formula One drivers of all time.
This data experiment was four years ago, though, and AI has come on leaps and bounds since then. Earlier this year, Abu Dhabi’s Yas Marina Circuit offered a glimpse of what the future could look like.
The Abu Dhabi Autonomous Racing League (A2RL) is a motorsport series with one key difference to the norm: there are no drivers in the cars.
Instead, teams must utilise their coding skills to programme the software in the cars so that these vehicles can perceive their surroundings, make decisions, and race competitively without human intervention.
An important caveat is that this a platform for research and development, so the on-track product should not be judged too harshly. This was evident at the first-ever race in April, which was punctuated by numerous issues.
The one-minute, four-second race highlights video on YouTube contains just ten seconds of footage of the cars actually on track. But this is but the first step on an exciting journey.
Crucially, that progress in just four years from a questionable list of the top Formula One drivers to AI being able to control physical cars out on a circuit is a strong representation of how rapid the rate of development has been.
18 months ago, no one saw generative AI coming, and yet every industry is now exploring how to best make use of the technology. That makes predicting the future difficult, but also means the possibilities for motorsport are truly endless.
This feature forms part of SportsPro’s AI Week in association with Stats Perform, a five-day run of exclusive content focusing on the sports properties and innovators driving sport’s intelligent revolution. You can hear more from the leading figures in this transformation and learn about the key themes at SportsPro AI, a two-day event at the London Stadium from 23rd to 24th September. Book your pass here.