Technology and Formula 1 have always gone hand in hand. The teams and their cars have been testing fields for new devices before arriving at street cars after a few years. But the broadcast has a new gadget to themselves thanks to artificial intelligence that is trying to help the audience understand what’s going on: F1 AI predictions. Whether you think they add an extra layer of excitement or you think it’s a spoiler… are these predictions right?
The Science Behind AI Predictions in F1
Formula 1 teams rely on a vast amount of data generated and collected by the cars, including tyre wear, fuel consumption, track conditions, and driver performance. The AI used by the broadcast with AWS is using a similar system, but using machine learning models instead of the calculations the teams use for themselves.
In the last few races, Formula One Management (FOM) has shown us different graphics with all this data. The highlights have been the tyre wear status, predicting when the tyres are going to give up on their drivers, and the possibility of a pitstop from a driver. The ‘catching-up’ infographic has received praise as well, as it’s helping the audience predict when a driver is going to be under threat of an overtaking from a rival.
However, there have been shadows since this technology was introduced that have put some doubts in the FOM about them, as they are not as accurate as they might seem.
While AI predictions in F1 can be quite impressive, there have been some circumstances that have created some confusion among the fans due to its mistakes. A driver’s race strategy can change at a moment’s notice due to an opportunity that has arisen for the team, making the predictions completely useless. A predicted pit stop might not happen if a team decides to take a gamble and extend a stint, or a driver might throw a curveball to their rivals by only one-stopping when the predicted strategy is a two-stop race.
This exact situation happened in the last Belgian Grand Prix. There, George Russell tried a one-stop race when both teams and FOM prediction was a two-stop race. Switching on the get-go to a one-stop took the British driver to an improbable race win, even though later the FIA disqualified him for driving underweight.
These are still pretty accurate if you take into account all the variabilities an F1 race can take. All in all, it’s a technology that is already present in other industries thanks to the shortcuts it gives to their processes. The casino games platform Jackpot City is already using it to recommend games to their users with a pretty good satisfaction rate so far.
The Role of AI in Shaping F1’s Future
Despite these occasional inaccuracies, the integration of AI predictions has been a great hit in enriching the viewer’s experience. Rather than watching cars circle the track, fans now have more context around race strategies, helping the most casual watchers understand better what’s happening in front of their eyes and the possible ideas the teams are discussing at the moment.
The fans now have to understand that AI predictions aren’t meant to be definitive, but only calculated guesses based on available data. As Formula 1 continues to adopt new technologies, future AI models should become more adaptive and take into account real-time variables.
