Industry response

A need to move faster

Our ER&D Pulse survey shows improving time-to-market is the sector’s top priority, coming out ahead of other priorities including cost reduction, capacity scaling, agility, and regulatory compliance. (Chart 1).

This urgency is rooted in executives’ lived experience. Three quarters say design timelines have increased over the last three years. Even more say production and times have lengthened, and 85% also say costs have risen (Chart 2). All this is happening whilst order books swell and backlogs mount.

AI and digital: Accelerating the engineering lifecycle

For energy and utilities, AI is an increasingly practical tool for delivering more infrastructure and maintaining more assets with limited engineering capacity.

AI to compress engineering and delivery timelines for energy assets

As energy companies expand infrastructure deployment, AI offers the opportunity to compress delivery timelines, enabling projects to be completed faster and at lower cost.

By applying AI to program and historical delivery data, companies can make more informed investment decisions and optimize asset development programs for new infrastructure and upgrades. AI can also help accelerate permitting processes, by rapidly gathering relevant information and responding to stakeholder and regulatory questions.

As projects get underway, AI can help to manage dependencies across multiple teams and identify emerging risks as they arise, helping teams intervene before delays cascade. Capital projects often involve multiple vendors working across different systems and ways of working. Recent advances in AI allow it to analyze these complex interdependencies and guide managers in coordinating activities towards a shared delivery goal.

What energy and utilities leaders think of AI *

expect AI to transform maintenance and support,
e.g. through predictive maintenance and decision support

expect it to transform documentation and compliance

say it will transform manufacturing and operations, e.g. through AI scheduling and supply chain management

*Data from Capgemini Engineering and R&D Pulse 2026^[2] survey of 200 energy and utilities leaders

Only 24% of energy and utilities leaders are currently using AI for engineering and R&D

Whilst not the most glamorous application of AI, it has already shown promise for significantly reducing documentation and compliance workloads^[3]. In fact, in Capgemini’s survey of 200 energy and utilities leaders – part of our cross-sector Engineering and R&D Pulse 2026^[4] report – we found 69% of energy leaders expect AI to transform this area, with potential to substantially shorten delivery times for major infrastructure programs.

For the engineers doing the work, AI assistants can automate document reviews, check compliance with standards, route approvals, and surface relevant information and contextual guidance. This can remove bottlenecks and enable quicker responses to changing project conditions. With limited pools of engineering talent, this reduces time spent on administration and troubleshooting, allowing engineers to focus on the areas where their expertise is most valuable.

Beyond accelerating capital delivery, AI is also set to become central to operating increasingly complex electricity systems. As renewable generation, batteries, EV charging, flexible industrial demand, and distributed energy resources proliferate, utilities will need to adapt. For example, integrating AI to optimize network flows, balance supply and demand in real time, and maximizing the use of both existing and new capacity.

AI for smarter asset management

Even as utilities build new infrastructure, they must also operate vast portfolios of both new and aging assets with limited engineering resources.

The combination of AI, digital twins, sensors, drones, and computer vision enables a shift from periodic, reactive maintenance to continuous, predictive asset management. These technologies can identify deteriorating asset conditions earlier, predict failures before they occur, and schedule maintenance, based on operational risk.

This is where energy leaders expect AI to have its greatest impact. Our research found that 75% believe maintenance and support will be transformed by AI over the next two to three years.

A real example of this potential comes from Capgemini’s work with EDP Redes España^[5]. By applying AI and computer vision to electricity network inspections, we reduced inspection times from a year to a month, while improving defect detection. The result was not only a dramatic acceleration in the inspection process, but also a significant increase in the organization's ability to monitor and maintain its asset base.

AI moves from the digital to the physical world

Perhaps most intriguingly, AI is moving beyond the digital realm to perform physical tasks in both asset delivery and maintenance. The convergence of AI and industrial robotics is creating machines that can perceive their environment, make decisions in unfamiliar or changing conditions, and undertake increasingly complex activities.

For energy companies, this opens new possibilities. For example, inspecting difficult-to-access assets, collecting operational data, and carrying out maintenance activities – particularly in hazardous environments where sending people can be dangerous, costly, and complex.

Consider the example of a nuclear facility. Accessing certain areas may require specialist personnel, detailed work planning, radiological risk assessments, air-fed personal protective equipment, contamination monitoring, and extensive decontamination procedures – before and after the task.

In a world of scarce specialist talent, deploying robots for some of these activities can free experts to oversee a much larger portfolio of work, while reducing operational risk. Capgemini has been testing industrial humanoid AI robots with nuclear fuel manufacturer Orano^[6], exploring how these technologies can support the next generation of asset operations and maintenance.

Although still at an early stage, physical AI and robotics have the potential to be transformative for asset deployment and management.

A need to move faster

Engineering at scale: Accessing the talent needed to build faster

The biggest barrier to faster infrastructure delivery is engineering capacity. Across the energy sector, organizations are being asked to design, build, and maintain more infrastructure than ever before. However, the supply of experienced engineers is struggling to keep pace. In many markets, these skills simply do not exist in sufficient numbers.

Leading organizations are therefore rethinking where engineering work is carried out. Rather than concentrating capability around individual assets, they are building global engineering networks that allow specialist expertise to be accessed wherever it is available.

Whilst an established way of working in industries such as automotive, global delivery has long seemed incompatible with large-scale critical infrastructure. Yet many engineering activities do not need to be undertaken next to the asset itself.

Engineering design, CAD, simulation, documentation, data management, digital twins, software and AI model development, and engineering analytics can all be delivered remotely. This allows local teams to focus on site-based activities, like stakeholder engagement, construction, commissioning, and regulatory approvals.

Ways energy and utilities leaders are addressing engineering and R&D capability issues *

outsourcing specific engineering or R&D activities

establishing owned offshore canters of excellence

partnering with external technology vendors, e.g. hyperscalers, platform vendors, innovation firms

*Capgemini Engineering and R&D Pulse 2026^[4]

What's more, the skills required to operate modern energy systems are evolving. Utilities increasingly need software engineers, AI specialists, data scientists, and cybersecurity experts to build and manage intelligent grids. Accessing these capabilities through global engineering networks can help address local talent shortages, improve cost efficiency, and accelerate digital transformation.

One useful way of thinking about this is through the Core/Context framework. This supports allocating work in ways that benefit from global talent, without creating unnecessary risk. Activities that differentiate an organization, such as those requiring local knowledge and relationships, or those involving protected intellectual property, should remain core to the business. More repeatable engineering activities can be delivered through specialist capability centers, where scale, standardization, and access to deep technical expertise improve both productivity and consistency.

As organizations adopt these models, careful consideration must be given to delivery location strategies. This can help to balance offshore and nearshore approaches to achieve the right mix of skills availability, cost efficiency, language requirements, and time-zone alignment.

The model is becoming increasingly mainstream. Capgemini's Pulse research found that 78% of energy leaders were outsourcing engineering or R&D work packages, while 73% were establishing offshore Centers of Excellence. The trend is also visible across the industry. Utilities increasingly invest in global capability centers to support infrastructure delivery and digital transformation across multiple geographies.

When combined with AI and digital engineering, global capability models offer the ability to scale engineering capacity more quickly than traditional recruitment alone. In a world where infrastructure demand is growing faster than the engineering workforce, that may become a significant competitive advantage.

Engineering capacity

A need to move faster

Organizational change:
Accelerating safely

AI and global delivery models hold huge potential, but for them to succeed, organizations must evolve in ways that make these capabilities part of everyday engineering practice.

For energy companies, that requires a careful balance. These organizations operate critical national infrastructure where safety, reliability, and security will always take precedence over speed. New ways of working, along with the adoption of AI and automation, must therefore earn trust, demonstrate value, and integrate with existing engineering disciplines – rather than seek to replace them.

Learn fast, scale fast

Successful organizations are unlikely to adopt AI or new operating models through large-scale transformation programs. Instead, organizations should identify well-defined operational problems and deploy AI in controlled environments to solve these problems, expanding their use as confidence grows. Each successful deployment creates evidence, builds organizational capability, and provides a foundation for tackling more complex use cases at greater scale.

A practical way to build confidence is to start with targeted operational use cases. Use cases like automating vegetation management or predicting equipment deterioration provide operational benefits without putting energy capacity at risk. They also give engineers hands-on experience of working alongside AI. Early deployments like these allow organizations to experiment, learn quickly, and scale successful approaches rapidly.

Along the way, organizations can refine governance, validate outputs and understand failure modes. They can also build familiarity, understanding, and trust amongst operators before applying AI to more critical functions. This staged approach allows AI to move quickly, but in a controlled and accountable way. For example, from assisting engineers with routine operational decisions to supporting (and eventually automating) carefully governed elements of control room operations.

The same principle of progressive adoption applies to people. Engineers do not need to become AI specialists overnight, but they do need to build confidence quickly in working alongside digital tools and understanding how AI arrives at its recommendations. Equally, leaders need to create an environment where experimentation is encouraged within clearly defined governance, safety, and regulatory frameworks.

Capgemini's work with EDP Redes España, as well as several other energy companies, illustrates this point. We have delivered programs that combine AI-enabled inspection and predictive asset management with structured change management. This is helping operational teams adopt new ways of working as new technologies are introduced.

These programs established common industrial data foundations, standardized operating processes, and robust governance. This allowed engineers to progressively incorporate AI into everyday decision-making, building trust while scaling technology across assets and into increasingly complex use cases. The result is not only improved inspection and predictive capabilities, but an organization that is better equipped to adopt future digital innovations.