Rolls-Royce reports 50%-plus order surge in data center energy as AI infrastructure demand reshapes power procurement

Rolls-Royce's data center energy business saw orders climb more than 50% in the first half of 2026, CFO Helen McCabe told CNBC on July 30. The figure, disclosed during a half-year earnings appearance on Squawk Box Europe, captures just how fast AI infrastructure spending is flowing to industrial power suppliers rather than staying inside the traditional utility or hyperscaler relationship.

The disclosure lands as energy operations teams across the renewables, grid, and battery sectors are absorbing a separate but related wave: AI-native software platforms designed specifically for energy workflows. Two announcements in the past 48 hours illustrate where procurement and R&D budgets are moving.

Rolls-Royce and the data center power equation

McCabe's comments, reported by CNBC, covered small modular nuclear reactors and data center power alongside the broader half-year results. The 50%-plus order growth figure is notable because it represents contracted pipeline, not revenue already booked, giving operations and procurement leaders at facilities-intensive organizations a leading indicator of where the industrial power supply chain is tightening.

Rolls-Royce has positioned its Power Systems division, which supplies high-reliability distributed generation, as a direct answer to the power density and uptime requirements of large AI compute clusters. The order surge suggests that data center developers are signing agreements well ahead of commissioning dates, a procurement behavior consistent with constrained supply and long lead times on industrial generation equipment.

McCabe also addressed UK investment in the CNBC interview, signaling that domestic energy infrastructure remains a strategic priority alongside the company's international data center push. For enterprise operators evaluating distributed generation partnerships, the combination of SMR development and near-term data center power contracts positions Rolls-Royce as a vendor active at multiple planning horizons simultaneously.

AI enters renewable asset management: the Quintas-Box deployment

On the software side, Quintas Energy, a global independent asset management platform for the renewable energy sector, announced July 30 that it has selected Box, Inc. as its Intelligent Content Management system. The deployment is the centerpiece of Quintas Energy's 2026 digital transformation program, according to a Business Wire announcement.

The operational rationale is straightforward. Renewable portfolios are growing in scale and complexity, and the documentation, contract, and sensor-data volumes that accompany wind, solar, and storage assets quickly outpace generic file storage. By standardizing on Box's ICM layer, Quintas Energy is building the structured content foundation that downstream AI applications need to ingest and act on portfolio data reliably.

For asset management teams at utilities or independent power producers evaluating similar infrastructure, the Quintas deployment is a reference case for sequencing: establish governed, AI-ready content architecture before deploying predictive or generative applications on top. The Box platform gives Quintas a single system of record across geographies and asset classes, which matters when the same documentation must satisfy multiple regulatory regimes and lender reporting requirements.

Battery R&D gets an AI-native toolset from Automat Solutions

Automat Solutions, headquartered in San Leandro, California, launched two products on July 29: LyteMatch and LyteGuide. The pair form what the company calls an Integrated Electrolyte Intelligence Workspace, targeting battery developers who need to identify and validate electrolyte formulations faster than conventional lab iteration allows, according to a Business Wire release.

LyteMatch addresses the formulation search problem, helping researchers narrow a large chemical search space to candidates with higher likelihood of meeting performance targets. LyteGuide adds a scientific Q&A interface, letting developers query the underlying research base without manually reviewing literature. Together, the tools are designed to compress the early-stage R&D cycle, which is where cost and time overruns tend to accumulate in battery development programs.

The launch is relevant to procurement and R&D directors at energy storage companies, EV manufacturers, and grid storage developers. Organizations that are evaluating electrolyte suppliers or running internal cell chemistry programs now have a commercial AI option for that stage of the workflow, one that integrates search and question-answering into a single environment rather than requiring researchers to stitch together separate tools.

What the convergence means for energy operations teams

Taken together, the Rolls-Royce order data and the two software launches describe a consistent pattern: AI demand is creating acute pressure on the physical power supply chain while simultaneously generating commercial AI tools purpose-built for energy workflows. The two dynamics are not separate stories for energy operators; they are the same operational challenge approached from different directions.

Facilities and procurement teams at data-intensive organizations need to be evaluating industrial power partners now, given the lead times implied by a 50%-plus order surge reported by CNBC. Asset managers at renewable platforms should be asking whether their content and data architecture is AI-ready before they commit to analytics or automation spending. And R&D organizations in battery chemistry are seeing the first commercially available, integrated AI workspaces arrive at a point when electrolyte innovation is a critical bottleneck for grid and EV storage scale-up.

Rolls-Royce's next formal update on the data center energy pipeline will be watched closely for whether the order growth has translated into revenue recognition and whether SMR timelines are firming. For Quintas Energy, the Box deployment is framed as a 2026 program, meaning integration milestones and early AI use-case results could surface by year-end.