Capgemini sees multi-year IT modernisation boom ahead of AI adoption

Capgemini Chief Executive Aiman Ezzat said companies hoping to deploy artificial intelligence at scale must first modernize decades-old technology systems, a shift that turns AI adoption into a multi-year rebuilding cycle rather than a quick burst of pilot spending. The work now being funded runs through application rewrites, data integration, cybersecurity, cloud migration and digital process redesign, all of which sit behind the flashy demos that have dominated the AI race.

Capgemini has spent the year framing itself as a beneficiary of that shift. In a February 2026 group presentation, the French consulting and technology services company said it helps clients move AI from small experiments into enterprise-wide production. On May 27, at its Capital Markets Day, Capgemini said it was poised to capture the full value of the "Agentic AI revolution" and laid out a 2028 strategic direction. Ezzat has also said legacy systems customized over years are no longer enabling innovation and are holding it back.

The numbers suggest clients are already paying for the cleanup. Capgemini reported first-quarter 2026 revenues of €5,943 million, up 7.0 percent at current exchange rates and 11.0 percent at constant exchange rates, and said it secured major transformational deals and long-term contracts. In the third quarter of 2025, revenues were €5,393 million, up 2.9 percent at constant exchange rates. The group’s full-year 2025 results were published on Feb. 13, 2026.

For Capgemini, which has 420,000 experts in more than 50 countries, the spending wave extends beyond a single product cycle. It points to demand for systems integration, managed services and software licensing as companies rebuild the data and infrastructure needed to run AI at scale. Older infrastructure vendors face a choice: adapt to that rebuild or lose ground as customers shift budgets toward cloud-ready, AI-compatible stacks.

The bigger commercial question is how long the payoff takes. Many firms spent the past two years exploring AI strategy; now the work is moving into implementation, where the cost of readiness can be as significant as the promise of productivity. Companies that still rely on fragmented or outdated systems may struggle to capture efficiency gains or deploy generative AI responsibly, making the first phase of the AI boom less about new software and more about paying for the systems that make the software usable.