Key takeaways
- AI will be the key to controlling grid complexity. Utilities that integrate AI into grid management will turn disruption into opportunity.
- Embracing change will unlock value. Utilities must evolve workforce cultures and forge strategic partnerships to stay agile and deliver returns on AI investment.
- Reliability starts with cybersecurity. As threats escalate, utilities that deploy AI-driven anomaly detection and automated responses will better protect critical infrastructure.
Over the last few years, the energy and utilities sector has experienced decades’ worth of change.
A system built for one-way power flow is expanding into a dynamic energy ecosystem. The data center boom is reversing years of stagnant electricity demand. And risks are growing, whether from extreme weather, evolving regulations, or AI-powered cyberattacks targeting the critical infrastructure that millions of people rely on every day.
Yet amid this upheaval, one expectation remains wholly unchanged: the delivery of reliable power. In the year ahead, how utilities meet this demand despite unprecedented disruption will constitute their most important — and challenging — test.
The biggest grid risk in the next decade isn’t weather or cyber-attacks — it’s organizational inertia.
Energy providers clinging to old production models will fall behind. Advanced AI technologies are transforming grid management and redefining roles across the sector, creating new opportunities to become more resilient, responsive and reliable while launching new services.
However, according to Kyndryl’s Readiness Report, nearly 70% of the sector’s leaders feel unprepared for external business risks — a significantly higher level of concern than those in other industries.
In 2026, organizations will need more than digital tools to steer through the sector’s wholesale transformation. Future readiness will require keeping up with an accelerated pace of change — one that shows no signs of slowing.
The biggest grid risk in the next decade isn’t weather or cyber-attacks — it’s organizational inertia.
Orchestrating a dynamic grid
Managing today’s complex grids requires AI assistance.
The predictable path of power flow from centralized coal or gas plants has been recast. Distributed energy resources (DERs) like solar panels and electric batteries can now generate, store and send power back to the grid, enabling greater flexibility. But existing grids were never designed for two-way power flow. Moreover, the output of these decentralized sources can be weather-dependent, and many have outdated monitoring systems that limit visibility and control.
To manage more flexible — but less predictable — grids, utilities are turning to AI-driven Distributed Energy Resource Management Systems (DERMS) to orchestrate diverse energy sources in real time. They’re also adopting enterprise-wide data lakes and unified platforms to integrate data across multiple domains, prioritize high-risk areas of the grids they manage and optimize operations.
Another significant opportunity lies in shifting from reactive to predictive maintenance. With edge devices like sensors, advanced data platforms and AI-enhanced digital twins that virtualize substations and grid segments, utilities can simulate grid behavior, anticipate failures, and simultaneously reduce outages and long-term maintenance costs.
Building an AI-ready workforce
AI is already transforming roles across the utilities sector. Field crews, for example, are starting to resemble data-driven problem solvers more than manual inspectors. Equipped with tablets, drones and AI tools for in-field analysis, these workers are now armed with digital tools to proactively fix issues and speed up repairs.
Nearly 90% of business leaders expect AI to transform their workforce in the next year, according to the Readiness Report. But a wide gap exists between this projected future and today’s reality. Many of the sector’s workers are not using AI, and the sector’s leaders are more likely to report inflexible workplace cultures that make it harder to adapt to change.
Utilities must evolve their cultures and equip their workforces to leverage emerging technologies. While utilities are prioritizing training and development, they are more likely than other industries to agree that their culture stifles innovation. And only 29% of the sector’s leaders — below the all-industry average — say adaptability and continuous transformation are core cultural values.
Overcoming cultural inertia will be critical for a sector that will need an AI-ready workforce to survive. As more utilities adopt autonomous agentic AI systems, only half report positive ROI on AI. To fully benefit from their AI investments, utilities will need to drive change — and shepherd their workforce through it.
Safeguarding critical systems
According to Readiness Report data, utilities are more likely than other industries to have experienced a cybersecurity breach in the last year, with more than a third reporting a major incident. In the EU alone, the number of attacks in the energy sector in the recent two-year period doubled. Global data points to significant growth in cyberattacks targeting utilities.
In part, this is because the convergence of IT and operational technology and the use of more third-party tools have expanded the digital attack surface. Today’s bad actors are quick to identify weaknesses in interconnected grid systems, networks and supply chains — and wield AI to carry out more damaging attacks. At the same time, underfunding has led to a buildup of technical debt and outdated systems that are harder to patch and secure.
The only way to stay ahead of AI-enabled attacks will be to fight back with AI. Indeed, nearly three-quarters of utilities are investing in AI-driven cybersecurity — more so than any other AI capability, according to Readiness Report data. These tools may include anomaly detection systems that instantly flag unusual patterns, as well as automated incident response systems that can isolate or shut down infected parts of a network. Tools like these that enable rapid action will be increasingly required to defend against fast-evolving threats.
Partnering for flexibility
On the demand side, utilities are exploring ways to incentivize how and when power is used.
Nowhere is this opportunity greater than with hyperscalers. For utilities, these server hubs — particularly those being expeditiously constructed to run AI workloads — are mega-loads, requiring as much electricity as mid-size cities. By 2030, global data center demand is projected to nearly double, according to S&P Global research.
But these data centers can also serve as grid stabilizers. Hyperscalers can shift workloads in real-time and use power electronics to help balance grid fluctuations. Indeed, emerging startups are now dedicated to turning data centers from energy strains into grid assets.
As interconnection backlogs grow, some regulators and grid operators are tying interconnection to flexibility. In Texas, for example, new rules mandate that large loads connecting to the Electric Reliability Council of Texas (ERCOT) grid participate in demand-response programs. Hyperscalers may also be able to connect faster by demonstrating flexibility. PJM, the largest North American grid operator, is weighing how to move forward with a new process to speed up interconnection requests.
However the future unfolds, utilities will need trusted partners at every turn to help them integrate technology and confidently navigate the tension between new models and traditional demands. Leaders’ ability to strike the right balance — one that helps to ensure the delivery of reliable power for years to come — will determine their success.
