With a legacy of over 130 years, the Michelin Group is best known as a global leader in tire innovation, but its ambitions go far beyond mobility. Headquartered in France and operating in over 170 countries, Michelin today is driving forward on three strategic pillars: people, planet, and profit.

From pioneering the first detachable tire to advancing sustainable materials and AI-powered services, the company is continuously evolving its identity as both a manufacturer and a technology-driven solutions provider.

In this three-part series, Ambica Rajagopal, Group Chief Data and AI Officer at Michelin, joins Julian Schirmer, Co-Founder of OAO, to unpack how Michelin is navigating strategic shifts, embedding AI across its operations, and preparing for a future where intelligent systems and sustainability go hand in hand.

In this first installment, Rajagopal outlines the company’s core business, its evolving priorities, and how AI is being used not only to optimize internal processes but to reshape customer and ecosystem engagement.

Edited Excerpts

Q: To begin with, what is the big strategic focus for Michelin right now? What challenge are you primarily addressing and how does AI play a role in this strategy?

Overall, our mission is to make our tires fully sustainable in the coming years and decades. We’ve also put together a set of dreams, frontiers that combine people, planet, and profit. So, people times planet times profit: we aim to balance all three aspects in every initiative we take. And our AI efforts are very much aligned with these goals.

For example, within our plants, we use AI to improve sustainability outcomes. One way is by monitoring deviations in the production process and reducing them — this results in raw material savings.

We’ve also used AI to speed up how we model the impact of design changes on tire performance. This has reduced the need for physical testing, and therefore, consumption of materials.

These are just some ways we’re using AI to improve sustainability, make jobs at Michelin more attractive, and increase process efficiency.

Q: How do you measure or communicate these sustainability gains, especially considering AI itself consumes energy? How do you balance both sides?

There’s been a lot of recent discussion about the CO2 impact of training large AI models. But having said that, most of the energy consumption happens during the training phase. What we’re doing — which is inference at scale — has a very long tail of usage, and that’s where we focus. There’s less conversation around that balance, but it’s important.

Without going into too much technical detail about how the CO2 impact is calculated, we firmly believe that monitoring and understanding usage is key for improvement.

As part of our infrastructure offering, we already monitor the CO2 impact of our AI models and compute usage.

Q: Outside of internal use, are there examples of AI-powered services at Michelin that directly impact customers or external stakeholders, maybe even public infrastructure, like using tire sensors to detect road damage?

Yes, absolutely. We have many AI applications deployed for external use. For instance, we have a chatbot on our UK website, and we are already seeing positive feedback from customers who engage with Michelin through it.

The example you mentioned — that’s part of the Michelin Mobility Intelligence services. We’ve built AI solutions that provide insights aimed at making roads safer.

For example, we’re able to give transportation departments insights into road quality and route risk levels. We are using AI not just for internal efficiency but also to offer value-added services externally.

Q: Acknowledging how quickly AI is evolving, what trends do you expect to shape the tire industry? Any parallels to how bots might soon be booking travel for us, as seen in other industries?

We definitely see a trend towards digital channels like websites, e-retail, and mobile apps — for both purchasing and engaging with tires.

Internally as well, we’re seeing major changes in how people interact with knowledge. We’re deploying AI agents and have broad adoption of generative AI to boost individual and team productivity.

At Michelin, we’ve already built a GenAI platform that keeps our data secure while offering access to these technologies, including specialized agents designed for our teams. It’s been amazing to see the curiosity and strong engagement this has generated.

We’re also exploring foundation models in areas where we’ve traditionally used AI, such as time-series forecasting. We now see promise in using transformer-based models instead of traditional statistical or boosting approaches.

Another exciting area is self-adaptive process control using reinforcement learning — particularly valuable in our manufacturing processes. That’s another promising field for us.

CDO Magazine appreciates Ambica Rajagopal for sharing her insights with our global community.