Inside Chip Ganassi Racing’s AI work on car setup and race strategy

Chip Ganassi Racing’s collaboration with OpenAI is developing around a concrete racing problem: helping engineers find the useful information for the next setup decision. A second episode of the partners’ R&D documentary, released on 7 October 2026, follows that work during Alex Palou’s 2026 championship campaign. The team’s account describes AI becoming more integrated into the work on car setup and strategy.[1]

For a racing team, a better answer can be valuable because it arrives while there is still time to try something. Practice offers a limited opportunity to compare a proposed change with what the driver actually feels and what the car records. Information from past sessions becomes more useful when it can inform that immediate choice.

A billion measurements still begin with a question

OpenAI says a Ganassi IndyCar carries more than 200 sensors and generates nearly one billion data points per hour during a race. Its description combines those signals with historical records, test sessions and pit-crew performance. The partners’ first year concentrated on making sense of the data; the next phase connects the resulting insights so engineers can devote more attention to decisions.[2]

The team’s October account names the ingredients engineers must reconcile: track conditions, tyres, aerodynamics, springs, dampers and anti-roll bars. These factors interact. A setting that helped in one context may answer a different question in another. AI-assisted retrieval and comparison can make the relevant evidence easier to reach, while the engineers interpret its significance for the car in front of them.[1]

The work reaches the pit lane and the conversation

The April partnership announcement describes three uses: race-data analysis, faster communication and support for decisions under pressure. It also says the No. 9 and No. 10 teams finished first and second in the 2025 Firestone Pit Stop Performance Championship, and attributes some operational efficiency to OpenAI tools. The placing is a team outcome; the announcement does not isolate the tools’ contribution.[3]

That wider scope matters. A race organisation can improve without changing the vehicle itself: a clearer comparison, a faster handover or a better understanding of repeated crew movements can all be useful. The common opportunity is to connect information across specialist activities that already exist.

Consider a hypothetical setup review. An engineer wants to compare two configurations used on similar tyre ages, then see whether the driver’s complaints changed at the same time as a sensor pattern. A useful system would help assemble those comparisons and surface mismatches in the conditions. That example illustrates the type of analytical assistance the reported collaboration makes possible; the publications do not disclose Ganassi’s precise implementation.

Learning the tools is part of the engineering

In an August interview published by the team, owner Chip Ganassi identified a practical constraint: the succession of race weekends leaves limited time to explore available tools. He said he wanted to use the offseason to improve his own understanding and expand the organisation’s use of OpenAI. Researcher Joyce Ruffell described weekly work with the team to determine where the technology could help.[4]

Ganassi’s new public account offers a useful view of AI adoption at that level. The interesting object is the developing working relationship: sensor streams and historical experience are being connected around decisions that engineers already make. The next advance to look for is a more specific account of a completed engineering task and the difference it made to the team’s work.

Report an error or suggest a correction