3. Cloud 3.0: All flavors of cloud
While cloud technology has been around for 15 years, there is a renewed momentum around the cloud, according to Georgia Smith, Capgemini UK’s Cloud Transformation Leader. As AI becomes the core workload, the cloud serves as both the execution layer and the environment for training models and running autonomous agents. There is not “one” cloud, but all types of clouds working together as one intelligent fabric.
This hybrid, multi-cloud environment is the natural progression of operationalizing AI, where agentic systems run cloud platforms. In this environment, organizations need to upskill constantly. If you’re only asking who the hyperscaler is that can meet your needs, you should also be asking if you have the team and the capability to run agentic systems.
4. The rise of intelligent ops
Simone Neser, AI Taskforce Program Manager at Capgemini Business Services, said that the rise of intelligent operations — hyper automated processes that combine data and AI to make processes more efficient and adaptive — signals a major transformation in how companies will harness the power of AI agents.
The fundamental daily activities that keep a business running are a great use case for AI in any organization. To build intelligent ops, you must decompose the process and understand each component before you can orchestrate multiple technologies, data sets, and types of work together. This is a complex process that requires a high level of understanding to transform legacy processes. In 2026, the expectation is to move far beyond the pilots of previous years toward a self-improving, agile system.
In 2026, organizations can’t get away with simply optimizing individual processes — intelligent ops requires optimizing end-to-end processes with integrated value chains that enable the management of both internal workflows and external partners. Humans and AI must work together, and this relationship will become more formalized and strategic, with AI focusing on high-volume, repetitive work, freeing humans to focus on higher-value work. For example, AI could handle invoice payments, allowing a human to manage tasks that require nuance, such as negotiating a contract or resolving a dispute.
5. The borderless paradox of tech sovereignty
Guillaume Renaud, Head of Cloud Transformation at Capgemini Invent France, said that the renewed sovereignty momentum is not an IT-only topic, considering the current state of geopolitical tensions and the fragile supply chain for chips, cloud, and AI infrastructure. Additionally, there is increasing scrutiny of where and how data is stored — especially given the risk of large-scale outages.
The “paradox” is that this tech sovereignty exists through interdependence. Renaud stated that total technological autonomy is an illusion, since chips, cloud, software, and AI rely on global supply chains, and that sovereignty is not about isolation but about strategic autonomy that is globally connected yet resilient, controllable, and shock-resistant.