How energy companies can best chart their GenAI journey

In my last blog, I outlined why embracing generative AI (GenAI) is key to enabling a successful energy transition, creating long-term value for energy companies. But how can companies determine how best to scale GenAI to suit their own circumstances?

This is the topic of many of my discussions with energy clients right now. Lots of organizations are experimenting (or beginning to experiment) with the technology, which is to be expected at this stage of the GenAI journey. But as companies get ready to implement the technology more broadly, we see challenges emerging in terms of how to scale successfully while balancing new risks.

Why some energy companies are establishing AI innovation hubs

GenAI and artificial intelligence (AI) are often talked about in the same breath, but they are not the same. GenAI has some unique characteristics that differentiate it from traditional AI.

GenAI can move from A to Z at speed with potential multiple quick wins. Because models come self-trained, value can be realized more quickly. GenAI’s multiple reusable components create greater synergies of investment and skills across use cases. The characteristics of the technology demand completely new skills and different monitoring and risk management. And, of course, GenAI is changing at lightning speed.

These differences are shaping how energy companies adopt and scale GenAI. Organizations with greater digital and AI maturity have been quick to recognize its huge potential while acknowledging that success will require new abilities and the capability to manage new risks. Leaders in these companies are expanding current digital operating models to accommodate the specificities of the technology or creating AI/GenAI centers of excellence or innovation hubs to rapidly grow the new skill sets and foundational enablers required.

EY teams helped one North American utility establish one of these hubs, assembling the key talent needed for a quick start and building a scalable, flexible and robust set of foundational AI/GenAI processes and reusable GenAI components. Together, we helped the utility company define and execute the hub operating model and AI strategy, establish AI governance and risk monitoring, develop common technical components for reuse and, ultimately, build a delivery pipeline for use cases.

The hub’s rapid prototyping environment moves use cases from concept to implementation at speed, enabling the company to prioritize business-led use cases with the potential to deliver value — and help accelerate time to realize this value.

We have seen other energy companies take a more tactical approach. They begin by experimenting with or piloting a few GenAI use cases to both learn about the technology while also laying the foundations for governance, safety, skills, and tech and data readiness. For these companies, it can be easier and less risky to start with internal use cases or those where data is readier to be consumed.

As these organizations progress on their journey, both the pain points and opportunities of scaling GenAI will begin to emerge, and they will eventually need to review their AI strategy and create their own hub or center of excellence, or expand existing digital “factories.”

Governance is a key enabler to scaling GenAI

Wherever energy companies are on the GenAI maturity spectrum, we see a common theme — a focus on building governance, which is a key enabler to scaling securely and successfully. 

The rapid acceleration of GenAI capabilities and the technology’s particular risks, including hallucinations and bias, as well as a changing regulatory environment, mean that any energy company’s GenAI journey must include a strong focus on ensuring trust in AI. We’re supporting several energy companies, including those at the beginning of their GenAI journey and those further along, to establish dedicated AI/GenAI governance frameworks.

For energy companies with less mature capabilities, it is important to start exploring GenAI in parallel with the establishment of AI/GenAI governance. This is partly to keep learning but also because experimenting with internal use cases with high business value potential allows for quick deployment once safety controls are in place.

Start sooner rather than later!

We know some energy companies are still taking a wait-and-see approach to scaling GenAI, hoping to learn from others, get a clearer view of costs and regulations, and perhaps avoid the risk of still-maturing technology. But the risks of this approach far outweigh any benefits. Energy companies that don’t act on GenAI now will miss out on quick productivity gains, fall behind in innovation, score lower in employee satisfaction, struggle to attract talent and fail to deliver on accelerated growth plans.

At a time when the energy sector is navigating unprecedented levels of disruption and volatility, companies simply cannot afford to ignore the transformative potential of GenAI. A year ago, my discussions with clients would be centered around the question: “Why should your organization prioritize GenAI?” Now I’m asking them: “What will happen to your organization if you do not?”

As GenAI capabilities increase, the penalties for inaction will only increase, as will the gap between leaders and laggards. This is why I always recommend that energy companies start their GenAI journey sooner rather than later! Experiment now. Score some quick wins to build those foundational capabilities to reduce risk, prove your ability to execute to realize value, and lay the path to scaling for success.

The views reflected in this article are the views of the author and do not necessarily reflect the views of the global EY organization or its member firms.