AI-driven data center demand has put utilities under the spotlight, making load growth a central issue in both media coverage and strategic planning. AI as a mega disruptor is forcing energy companies to reimagine how they build and maintain infrastructure, partner with hyperscalers, manage peak load times, and more to ensure they can maintain reliability and affordability amidst the AI boom.
But that’s only half the story.
Utilities are not immune to the challenge of implementing – and extracting value from – AI as a means to evolve their own business operations. Last year a Gartner study revealed that 94% of energy company CIOs were planning to increase investments in AI. This is not surprising given the constant pressure to keep up with the pace of AI advancements.
However, investing in AI does not automatically result in value. In reality, the risk of wasted investment is pretty staggering. MIT found that only 5% of investments in AI were yielding meaningful return and that the energy and materials industries are seeing the lowest adoption and experimentation rates.
So, how can utility leaders manage the tension between the need to keep up with limited success to date?
Flip the Script on Your Strategy
First, energy leaders need to flip the script. Stop asking, “What’s our AI strategy?” and start asking, “How and where can AI help accelerate our business strategy?” If you start by trying to define your AI strategy, you’re likely going to have a very limited lens on what’s possible, and the risk of developing generic AI mandates skyrockets. We see this across many of the utilities we work with because of the pressure to prove the workforce is embracing AI. Leaders put out mandates like, “We expect everyone to use AI at least once a week.” When the goal is simply usage, people tend to use AI for basic tasks like taking meeting notes. While this might streamline some manual tasks, it is not generating the type of return that justifies even the initial investment in an enterprise-wide AI tool.
To create clarity of direction and increase connection to concrete outcomes, think about how AI can help in achieving your broader business strategy at pace. Are there ways AI can increase reliability for customers? Will finding operational efficiencies through AI help mitigate the risk of labor shortages, recognizing the challenge facing the industry of backfilling anticipated retirements? Can AI augment call center staff to improve customer experience and increase satisfaction scores?
First, energy leaders need to flip the script. Stop asking, “What’s our AI strategy?” and start asking, “How and where can AI help accelerate our business strategy?”
Reimagine Your Execution Plan
Once you have a vision for how AI can accelerate the achievement of your business strategy, it’s critical to set up two parallel pathways for execution: broad experimentation and targeted use cases.
Broad Experimentation
Unlike most digital tools, like SAP, Oracle, Maximo, etc., there is still so much that is unknown – and rapidly evolving – around what is possible with AI. There’s no standard playbook of best practices on how to optimize AI tools or drive adoption. What we do know is that:
- AI is rapidly evolving, with new capabilities coming to market constantly, which presents challenges and opportunities
- There is a deep sense of anxiety across many in the workforce, driven by a host of factors (fear of being left behind, worry about job loss, trepidation about skillsets, etc.), that much be addressed
- It’s nearly guaranteed people across the organization are already experimenting with AI, even if they aren’t advertising it
Given the number of unknowns and the needed mindset and behavioral shifts necessary to capitalize on AI use, leaders need to create safe spaces for experimentation. Provide clarity on what the organization is looking to achieve, and then let people test and learn. There are a number of ways to start doing this:
- Create discrete moments dedicated to learning and getting comfortable with new tools. One client did this by hosting a “Prompt-a-thon” that helped people better understand how to effectively write prompts and train their internal AI agent.
- Activate early adopters to serve as peer-to-peer catalyst. Identify those in the organization that are already using AI effectively and ask them to help educate and support their peers.
- Amplify the wins. And not just quantifiable wins like, “I saved 3 hours by…” Also highlight new behaviors, so people understand it’s safe to test and iterate.
Not all experiments will be successful, but there will inevitably be wins that can then be intentionally scaled more broadly across the organization. And building a muscle around experimentation and appropriate risk-taking will serve the organization well in the future, even outside the push around AI.
Targeted Use Cases
While there’s a lot we are still learning about AI tools, there are likely some areas with a very strong case for AI adoption. In general, processes that are repetitive and/or data heavy are ripe for finding efficiencies. Things like streamlining month end close through automation or using predictive analytics to set daily routes (and work bundling) for field workers are likely to have significant, measurable benefits to the business.
Once you have identified these use cases, tied to your strategic priorities, create opportunities for those closest to the work to craft solutions. Consider pulling together cross functional teams, made up of all the functions represented in the end-to-end workflow. For example, if you’re looking to leverage predictive AI to determine daily routes, you’ll likely need to pull a team together of representatives from the field, customer service, workforce planning, technology/IT, etc. Executives should set the charter and outline the challenge, opportunity, guardrails, and timeline (think short timeframes… 90 to 120 days). Then, unleash the team. Let them set the specific goal, determine the activities that are needed to achieve that goal, and give them space to experiment along the way. Leaders should be there to remove barriers and, when needed, help keep the team motivated to keep pace.
These targeted sprints can help others across the business see what’s possible by engineering meaningful, quantifiable outcomes at speed. Paired with broad experimentation, utilities are much more likely to see real value from their AI investments.
Leading the Future
Utilities don’t have the luxury to wait for others to be first movers when it comes to AI adoption. The entire industry is being disrupted, including internal operations and external demands. The time is now for energy leaders to shift gears and pivot how they approach AI adoption and value creation. The door is wide open for utility leaders to power the future of AI. Are you ready to walk through it?
