When Formula One returns under the lights of the Las Vegas Strip on Nov. 21, Oracle Red Bull Racing will already have run the race billions of times.

The cars start in a relatively quiet corner of the city — a former parking lot — before they loop around the Sphere, turn onto the Strip near the Venetian and, at upward of 220 mph, race past Caesars Palace, the Bellagio fountains and Paris Las Vegas. Then they do it 49 more times.

Long before the first of those laps, the team behind Max Verstappen, who won here last year, has modeled how they might unfold. It runs 8 billion Monte Carlo simulations before each race weekend, then another 8 billion once practice data comes in. Monte Carlo methods use repeated random sampling to estimate the probability of different outcomes.

“A race weekend is a data-driven event,” Jack Harington, head of technical performance partnerships, said at Oracle AI World in Las Vegas in late October. The team behind the driver Max Verstappen collects around 2 terabytes of data per car for every on-track session.

He said moving the simulations to Oracle Cloud Infrastructure (OCI), the cloud platform of the team’s title sponsor, has allowed the team to test how rivals might respond, too. “We’re able to play war games,” Harington said. “Since 2021, every winning decision we’ve made on strategy has been made on OCI.”

Red Bull is not along. Nearly every F1 now has a technology partner, with Microsoft, AWS and Google among the companies involved.

This season’s new regulations have raised the stakes. Oracle says OCI lets the team run more granular simulations that account for energy usage, active aerodynamic configurations, deployment windows and tire interactions across thousands of potential race scenarios.

Racing with a recommendation agent

The team has also put an AI agent on the pit wall with Oracle.

Engineers load sporting and technical regulations, and past decisions, into a set of LLMs. The tool retrieves the relevant rule and can offer a decision, which an engineer reviews.

“In a race, you don’t want an engineer leafing through documents,” Harington said. “You want them focusing on what Max is doing on track.”

Building an engine in the cloud

Red Bull Ford Powertrains, the team’s in-house engine operation, designed and tested its first hybrid power unit on OCI. The engine debuted this season.

According to Oracle Red Bull Racing, the team used OCI’s high-performance computing, bare metal infrastructure and large-scale simulation capabilities to design, validate and refine the engine in record time. Updates and integration work will continue on OCI through the season.

Away from the track, the team plans to use Oracle Fusion Cloud Applications for finance, HR and marketing, including a new fan loyalty program.

Finding the next driver

Next, the team wants to apply AI to driver scouting, a project still in development. “Historically, you might just look at the obvious national car racing series,” Harington said. The team plans to widen the search to “find diamonds in the rough that other people haven’t identified.”

Boats as IoT devices

SailGP, the league launched in 2018 in which sailing teams race identical F50 catamarans, uses data a little differently than F1. Teams can see one another’s numbers through Oracle-powered dashboards, and the league uses the sensor data for maintenance.

“Our F50s are like extreme IoT devices on water,” Matt Lord, director of technology at SailGP, said at the same event. Because the league builds and maintains the boats, “there’s nothing worse for a team, a sponsor or the league than a boat that can’t make it to the race and onto the water.”

An anomaly detection program flags components that could fail. “At probably every race, we find one or two things that are an area of concern, and we switch out the parts,” Lord said.

The league is now building an AI assistant for coaches and athletes that can explain, for example, why one boat went left and another right. It wants to bring similar tools to broadcasts in more than 200 territories, alongside its fan-facing data product, SailGP Insights, which Lord called “a true second screen.”

“If we can demystify it, we can bring more fans to the sport,” Lord said.

Personalized Feed
Personalized Feed