By Christopher Piccolo, HEXstream utilities industry specialist

Last week I attended UA Week 2025 in Austin, Texas. The event brought together analytics and data professionals from utilities across North America and highlighted one clear message: AI is now at the center of every conversation about utility analytics.

AI Is driving analytics everywhere

Across multiple sessions at UA Week, including sessions such as Generative AI, Analytics Architecture & Technology, Asset Health Analytics, and Safety Analytics, the shift toward AI-first analytics was undeniable.

In asset health and grid analytics, AI is being used to interpret large volumes of data (such as drone imagery during storms), helping teams assess damage and plan restoration faster. In IT and operations, AI-driven monitoring, anomaly detection, and process automation are becoming common. Some utilities are now using AI agents connected to internal data sources to accelerate productivity by analyzing safety manuals, regulatory guidelines, and policy documents.

The message was clear: AI is being woven into every layer of the utility enterprise.

The crawl-walk-run philosophy

One of the most repeated themes across the breakout sessions at UA Week was the idea of maturity. Many utilities are still learning to crawl when it comes to AI adoption, with plans to walk and eventually run as their data and processes mature. Organizations are starting with small pilot projects, building internal data literacy, and developing governance frameworks before scaling broadly.

The message resonated: moving carefully is critical to success.

Data quality as the foundation

Another strong theme throughout the agenda was that AI can only deliver value if the data behind it is trustworthy. In the analytics and enterprise-data sessions, speakers emphasized the importance of strict data governance, data-quality frameworks, standardized models, and defined stewardship roles. Without consistent and accurate data, even the most advanced AI systems struggle to produce meaningful results.

Many utilities shared that their current focus is improving data quality and governance as a necessary first step before expanding AI use cases.

Key session highlights

• The Generative AI track explored how utilities are using AI not just for prediction but for generation, including automating guideline-creation and summarizing policy content.
• The Asset Health Analytics track demonstrated how AI can prioritize asset maintenance by analyzing sensor data, maintenance logs, and inspection images.
• The Safety Analytics sessions focused on using near-miss and observation data to reduce risks and improve field safety through predictive models.
• A major general session outlined how utilities are evolving from descriptive dashboards to prescriptive along with automated decision-making systems, following the crawl-walk-run progression.

Together, these discussions showed that UA Week 2025 was not just about analytics tools, but about the transformation toward AI-driven decision making across the utility enterprise.

Final thoughts

UA Week 2025 made it clear that the conversation in utilities has shifted. The question is no longer whether AI belongs, but how to implement it responsibly, securely and effectively. With strong data foundations, clear governance, and a maturity mindset, utilities can unlock the real potential of AI to improve reliability, safety and operational excellence.