AI in energy: Cross-sector strategies reshaping real-world operations

In our latest Energy Transition Talks episode, Peter Warren sits down with CGI experts Diane Gutiw, and Lukas Krappmann to explore how artificial intelligence is reshaping the Energy & Utilities sector. This second instalment of their conversation covers trends from the recent Hanover Messe event, the challenges energy providers face and how cross-industry insights are accelerating innovation.

Catch up on part one of their discussion here.

How AI is addressing workforce and knowledge gaps

Energy providers today are under strain from all directions. Budgets are tightening. Retirements are accelerating. Organizations are under pressure to do more with fewer people while maintaining complex systems. AI is increasingly being applied to fill these gaps, enabling companies to automate routine tasks, retain knowledge and help employees focus on high-value work.

“There are tools that are replacing tasks, not people… so that people are able to do what they need to do to be creative.”— Diane Gutiw

Optimizing energy systems with real-time data and hydrogen

As technologies mature, the merging of physical infrastructure and digital intelligence is accelerating. From lidar to radar to industrial sensors, energy systems are generating more data than ever. AI tools are helping organizations interpret this data—along with its metadata—to make faster, better-informed decisions. In many cases, generative AI is also being used to correct and validate data in real time, improving quality and trust.

Hydrogen is emerging as a key use case for AI-enabled decision-making. Organizations are applying digital twins to optimize hydrogen production, integrate electrolyzer data with broader grid infrastructure and manage maintenance cycles. These systems help determine when to generate hydrogen, forecast energy needs and respond to real-time inputs like weather and demand.

“There’s a lot of data out there—but the challenge is what you do with it.”—Lukas Krappmann

Bringing cross-sector AI strategies into energy

Industries are no longer solving problems in silos. The energy sector is rapidly adopting tools and techniques originally developed in areas like manufacturing and healthcare. Shared challenges—like aging workforces, complex asset maintenance and efficiency demands—are driving knowledge transfer across ecosystems.

Predictive maintenance is one example where proven models from drilling operations are now being used in utilities, mining and even hospitals. By reusing patterns and data science approaches, companies are solving problems faster and more cost-effectively.

“It’s a best practice in data science… not to reinvent the wheel, but to find efficiencies and reuse patterns.” — Diane Gutiw

Responding to regulatory and supply chain disruption with AI

In today’s volatile environment, decisions that once took years now must be made in days. Shifting supply chains, dynamic regulations and volatile tariffs require flexible, AI-powered tools that help organizations adapt quickly. AI is increasingly seen as essential to achieving that agility, enabling organizations to model outcomes, manage regulatory requirements and optimize operations “on the fly.”

At a global level, macroeconomic and geopolitical shifts are also pushing traditionally separate sectors—like energy, logistics and life sciences—to innovate together. AI is emerging as a key enabler of this collaboration and innovation across industries.

Three ways AI is reshaping energy operations

The current wave of AI-driven transformation is coalescing around three high-impact focus areas:

1. Tools for internal efficiency: Enhancing access to data, automating administrative workflows and reducing time-to-decision
2. Advanced automation: Enabling more autonomous systems capable of handling complex tasks without predefined instructions
3. Data intelligence: Mining large, diverse datasets to uncover patterns and power evidence-based decisions

As energy organizations continue to digitize, AI is emerging not just as a support function—but as a foundation for more resilient, adaptive and data-driven operations.

Listen to other podcasts in this series to learn more about the energy transition

Read the transcript

1. Reintroducing our experts and their roles in AI and energy

Peter Warren

Hello everyone and welcome back to part two of our discussion on AI in energy and utilities and how this is impacting the energy market. It's part of our ongoing energy transition conversation. In part one, Diane and Lukas talked about what a digital twin is, digital triplets and how it's being applied. Today, we're going to pick up a few more things on innovation and the impact of it. But why don't we start with a reintroduction of yourselves. Diane? Do you want to go first

Diane Gutiw

Great, hi, Peter. Thanks for having me back. My name is Diane Gutiw. I'm a Vice-President at CGI and I lead our global AI research center. Over to you, Lukas.

Lukas Krappman

Yeah, thanks, Diane, for the introduction. My name is Lukas Krappman. I'm from Germany and therefore responsible for all of our manufacturing, or more the Haltrim business there. During the past, we're up with a couple of companies and trying to actually generate more data from the value of the product.

2. Industry trends and pressures influencing AI adoption

Peter Warren

Thanks very much. Bringing in some information from our ongoing surveys with customers and others in the industry—our Voice of the Client survey—we pointed out this year that, and this is not fully published yet, there's a lot of people interested in AI. They're interested in automation. They're trying to do more with less, manage not being able to get all the people they want, deal with cutting budgets, and account for retiring employees with key knowledge. One of our customers said, "Everything's happening everywhere all at once," which I think is a movie quote. Diane, what's your thought on the most exciting trends for energy and transitions? How do you see this AI actually being able to fill those problems I just described?

Diane Gutiw

I think we saw a lot in Hanover. I know we were all just at the Hanover Messe event and there is so much going on. It's hard to nail down one thing. Automation and the move to agentic AI to extend what we were already doing in intelligent automation into agentic AI is brilliant. Also, the efficiencies that organizations are gaining through use of some of these general purpose tools.

Peter Warren

And Lukas, we also saw the confluence between the logical layer and the physical layer—devices like your clients are using—and the overlap into physical security, use of imagery, lidar, radar, all bringing a bigger holistic viewpoint. What was the most exciting thing we picked up at Hanover and where do you see your customers going?

Lukas Krappman

What was interesting from my perspective was that everything's basically available now in terms of technical knowledge and tools. The macroeconomic challenges—how everything comes together and how the whole ecosystem of manufacturers, energy companies, logistics companies and even health and life sciences—can really work together to solve pressing problems. Especially due to geopolitical and economic influences, this is the most interesting thing for our clients right now.

3. Complexity, automation and AI-enabled decision-making

Peter Warren

You bring up a good point that supply chains and ecosystems are under attack. What used to take two years now must happen in two days. We've also heard clients want better software tools to be more agile in regulatory and tariff management. Diane, how do you see AI playing a role in all of this complexity?

Diane Gutiw

We're at a good place with tools that allow us to do what used to be really complex and expensive. Beyond the economic climate, energy is at the forefront and we also have a rapidly aging global population. Across all industries, there's a challenge of doing more, and more specialized work, with fewer people. Deep knowledge is retiring out. AI can help maintain and retain knowledge, upskill faster and take over repetitive tasks so people focus on higher-value work. Tools aren’t replacing people, but tasks—making operations more efficient and creative.

4. Cross-industry insights and transferable AI strategies

Peter Warren

If I can ask an open-ended question: we've hit on cross-industry parallels. At CGI, we see problems in one industry that are solved already in another. Can either of you comment on how we're learning across sectors?

Diane Gutiw

Sure. I spend a lot of time across industries and I think there is better collaboration now. At CGI, we support specialists across sectors for that reason. In data science, it’s best practice to reuse patterns. Predictive maintenance is a good example—we’ve applied models used in drilling to mining, utilities and even healthcare. Problems like an aging workforce are cross-industry. Efficiencies can be shared and vendor partners also help spread innovation across sectors. That’s a key value of our team at CGI.

5. Digital twins, triplets and cross-sector data ecosystems

Peter Warren

Lukas, at Hanover Messe you ran a workshop on digital twins and triplets. What did you take away from that? What were the clients most interested in?

Lukas Krappman

It was a very interesting workshop. We structured it around digital twins and triplets in the hydrogen ecosystem, but participants came from various industries—hydrogen, manufacturing, agriculture. What they had in common was the need to make data available quickly and extract insights. For example, in agriculture, connecting field data to infrastructure and business applications to optimize water use. In energy and utilities, hydrogen clients needed to integrate data from electrolyzers into large grids, do predictive maintenance and make informed production decisions. The challenge is connecting all the data and metadata to make decisions. That’s going to be a real focus in the next few years.

6. Where AI is heading: Real-world adoption and future priorities

Peter Warren

Thanks, Lukas. It’s interesting how a hydrogen use case applied to agriculture. As we wrap up, what’s your vision for where this is all going globally? Lukas, let’s start with you.

Lukas Krappman

Walking around the booths at Hanover, AI was everywhere. Every company had AI built into their business model—from intelligent copilots to custom GPTs. From what we heard from clients, it’s all about how AI improves daily efficiency and delivers real benefits. That’s the goal.

Diane Gutiw

Yeah, you know, I think the thing that's happened in the last year is AI now isn't just one thing; it's not just generative AI that people have on their phones. I find that innovations are splitting into three things, and those three things are really what we saw in Hanover, as Lucas said, as well as what's going to drive us forward.

The first is tools for internal efficiencies. This is your operational, administrative access to information, access to data to be able to do things in your day-to-day workflow quicker. So, a lot of what came out of the GPTs and software development and being able to integrate with everyday work tools is going to be a huge impact on every sector, but will be able to help a lot.

The second would be automation. We already see a lot of automation in this sector, with lots of robotics and automation and process engineering being advancing technologies in this space. I think that by having solutions that can do more, having more autonomous automation, having automation that's very focused on more complex tasks which were hard to complete without clear instructions, you're able to do more.

And then the last one is data. We've talked about data a lot, but using AI to be able to mine that data, to find patterns and have more evidence-based answers to questions. You know, to me those three things are splitting out of this technological evolution that we've seen and those are, I think, what the three top areas that are going to have a real positive impact on the energy sector.

Peter Warren

I’ll add one last thought from another conversation we had, Diane—using generative AI for error correction and data validation in real time. That’s a big leap forward. Thank you both. Hopefully everyone enjoyed this session. We’ll have more coming up in the weeks ahead. Thank you, Lukas.