Powering Through Disruptions: How AI Is Transforming Customer Support for Utilities
Chris Adomaitis, Director of Solutions Consulting, Omilia
Outages, rising operational costs, and an escalating wave of fraud attempts continue to put pressure on the utility sector. To stay resilient, many providers are turning to AI and advanced automation capabilities to streamline service, strengthen security, and maintain compliance even during their busiest outage seasons.
AI-driven systems can quickly sort and resolve customer inquiries in real time, cutting wait times and reducing dropped calls. Voice biometrics and deepfake detection technologies are bolstering security, while natural language understanding allows customers to describe their needs in their own words. No more endless touch-tone menus or confusing routing menus. The result is an engagement model built for today’s always-on customers - one that is faster, smarter, and far more resilient than legacy systems.
When the Lights Go Out, Customers Need Connection
Few experiences are more universal than suddenly losing power. The lights dim, the Wi-Fi drops, and our first instinct is to grab our phones for updates or reassurance. In 2023, U.S. electricity customers averaged over six hours without power, according to the U.S. Energy Information Administration. As inquiries pour in and frustration mounts, customers expect instant, honest communication. Contact centers are put under immense pressure to deliver clear, immediate information.
Historically, utilities have met this challenge with a human-first approach; pulling in extra staff or extending shifts during major disruptions. Operating with a reactive model, however, has become increasingly unsustainable. According to Salesforce, 88% of service professionals reported that customer expectations are higher than they used to be, and 82% said customers are asking for more than before. Overworked and overwhelmed workers can result in longer queues, incomplete resolutions, and frustrated customers.
The problem has been documented across the country. In Michigan, regulators discovered that DTE Energy’s call-center systems were severely overwhelmed during the winter storms in 2023. Nearly 44% of all customer calls were abandoned before reaching an agent and outages in the system left calls stuck in queues even after the caller hung up. These failures increased frustration, slowed down restoration efforts, and strained operations. These incidents underscore a critical point: utility communication is no longer just a service issue - it is a public confidence issue.
Combining Automation with Human Support
To meet rising customer expectations, utilities are using AI tools that extend their human workforce. Intelligent virtual agents can handle routine, high-volume inquiries like checking outage statuses, processing payments, or reporting issues, all while directing complex or sensitive cases to live agents. A hybrid model of virtual and human agents is quickly becoming the industry standard.
When an escalation happens, the AI system transfers a complete conversation summary and call history, so customers do not have to repeat themselves. This handoff helps agents resolve issues faster and allows them to focus on interactions that need empathy or discretion. Smarter, context-aware collaboration between people and machines underscores the value of the AI system that extends beyond simple automation.
Natural language processing enables customers to engage with the AI system in a conversational way. It removes rigid Dual-Tone Multi-Frequency (DTMF) menus and repetitive “press 1 for billing” prompts. Customers can simply say what they need, such as “I want to report an outage” or “check my bill,” and the AI responds instantly. By removing friction, utilities can boost the use of automated self-service while keeping a personalized experience. The payoff is significant: higher containment, higher CSAT, and reduced agent load - all without sacrificing human connection.
Using AI to Strengthen Security
The rise of Generative AI has created new opportunities for fraud through voice phishing and deepfake technology, a rising concern for utilities. Criminals can now use AI to access accounts without permission and compromise contact centers. To combat these growing threats, utility companies are using a layered defense. This includes voice biometrics to confirm a caller’s identity, detection methods for spoofing and deepfakes to identify fake audio, recognition of replayed or synthetic voices, and dynamic watchlists that monitor known fraud patterns and repeat offenders. As attacks become more sophisticated, these defenses are shifting from “nice to have” to mission-critical infrastructure. A built in, multi-layered defense boosts security and makes authentication easier, improving efficiency while protecting customer data and company systems.
Balancing Efficiency, Cost, and Compliance
Moving customers to natural language self-service requires breaking old habits and addressing skepticism from those who are used to traditional touch-tone systems. To encourage adoption, utility providers should focus on what customers care about most: quick, accurate information that leads to immediate solutions and easy transfer to a human agent when necessary.
Financial pressure adds another challenge. Contact center budgets at many utilities have decreased just as service demands are increasing. For agents, smaller budgets mean they have to do more with less, often while handling spikes in call volume during outages. Sentiment analysis tools can identify frustration or distress in a customer’s tone, which allows for a smooth transfer to a human agent when empathy and reassurance are needed. When staff are able to concentrate on sensitive or emotionally charged situations while virtual agents handle the repetitive inquiries, the workload becomes more manageable and can directly reduce burnout in an industry that is already facing staffing shortages.
Meanwhile, the cost of service underperformance continues to rise. In Michigan, regulators have set the stage for utilities like DTE Energy and Consumers Energy to face up to $10 million in annual penalties for missing reliability and service-communication targets. It’s clear that failing to meet communication and service standards can directly impact a utility’s bottom line. In this regulatory environment, it’s not good enough for AI to purely be a tool for efficiency. It’s essential that it acts as a safeguard against operational and financial risk.
Supporting the Future of Utility Support
AI-driven customer engagement is now essential for maintaining agile and compliant operations. By improving first-contact resolution, simplifying authentication, and speeding up access to vital information, AI helps utilities meet regulatory standards and provide reliable service, even during crises. Ultimately, the utilities that will lead the next decade are those that treat AI not as an add-on, but as a core component of their service strategy. As the industry continues to move forward, AI will be the foundation of customer communication. This will make utility support quicker, more personal, and stronger when it’s needed the most.