C3 AI and Shell have expanded their collaboration across Shell's global operations, extending a multi-year reliability programme.
Shell will expand its use of C3 AI's reliability software beyond anomaly detection by adding AI agent-based root cause analysis and remediation across its asset operations on Microsoft Azure. Shell's existing predictive maintenance programme already monitors more than 13,000 pieces of equipment.
The move deepens a relationship that began in 2018, when C3 AI started working with Shell on predictive maintenance at industrial scale. The programme was designed to identify potential equipment issues before failure and reduce unplanned downtime across energy infrastructure.
Under the expanded arrangement, the system will not only flag anomalies in equipment performance but also investigate likely causes and support remedial action. The deployment is intended to cover Shell's global asset base, extending the operational use of AI in heavy industry.
The collaboration also brings Microsoft further into the project through Azure, which hosts the software used for the reliability programme. The cloud platform serves as the foundation for running the system across Shell's operations.
Industrial use
The announcement comes as many AI deployments remain focused on software assistants and office tools, while industrial groups test how the technology can be applied in physical operations, where equipment failures can carry major cost and safety implications. Predictive maintenance has been one of the more established applications, but root cause analysis and remediation push AI into a more active operational role.
For Shell, the existing maintenance programme has already delivered significant value. C3 AI said the work has generated hundreds of millions of dollars in economic value by reducing disruption and improving reliability across monitored assets.
That places the partnership among a relatively small group of industrial AI deployments operating at large scale in production settings. Energy companies have explored digital tools for years, but broad rollouts across thousands of pieces of equipment remain difficult because of the complexity of plant operations, data integration, and the need for reliable decision support.
Stephen Ehikian, President of C3 AI, described the company's view of the expanded deal.
"C3 AI is the leader in industrial AI, and this expanded partnership with Shell proves what's possible when Enterprise AI is fully operationalized at global scale for predictive maintenance - reducing unplanned downtime and delivering hundreds of millions of dollars in economic value. Shell has built mature AI predictive maintenance programs on our platform, and together we're now pushing into agentic AI, advancing how this technology can further transform reliability, safety, efficiency, and operational performance," Ehikian said.
Cloud role
Microsoft also presented the project as an example of AI applications moving from pilot schemes into day-to-day industrial use. Azure underpins both C3 AI Reliability and the C3 Agentic AI Platform used in the Shell deployment.
Sandy Gupta, VP GISV, Software Development Companies at Microsoft, said the programme shows how AI software and cloud infrastructure can be combined in a live operating environment.
"What Shell and C3 AI have built on Azure over the past several years is exactly what enterprise AI should look like - real applications, running in production, delivering measurable value at global scale. This deepened collaboration is a powerful proof point for what's possible when world-class AI applications meet trusted, secure cloud infrastructure," Gupta said.
The expanded agreement shows how industrial groups are trying to move AI from monitoring systems into tools that can support diagnosis and action. In sectors such as energy, where outages and equipment failures can be expensive, the commercial case for that shift depends on whether software can improve reliability without adding operational risk.
Shell's programme now stands as one of the larger examples of AI-assisted maintenance in industrial operations, with more than 13,000 pieces of equipment under monitoring across its global asset base.