Electric utilities
From asset information and network planning to customer service and regulatory evidence, explore the work AI can support.
OUR MISSION
Helping utilities make the AI journey.
The people who keep essential services running deserve a clear path through a changing world.
AI for Utilities exists to make that path easier to understand: where AI can help, what a sound foundation looks like, and how to turn an opportunity into useful work.
Explore the utility AI journey
WHY WE’RE HERE
A more informed utility sector. A more practical way forward.
Utilities face an extraordinary range of AI possibilities. Finding a good starting point takes more than a list of technologies. It takes an understanding of the people, information, and decisions behind everyday work.
We bring practical guides, use cases, and planning tools together in one open resource. Our aim is to help teams ask better questions, make informed choices, and build capability one useful step at a time.
The guidance is for leaders setting direction, practitioners improving a workflow, and the technology and data teams helping make it happen.
WHO WE SERVE
Different services. Shared questions.
For the people working across electric, gas, and water utilities.
Gas utilities
Apply useful patterns to documentation, field work, customer support, and planning—with the data and controls your service requires.
Water utilities
Connect asset knowledge, inspection records, customer needs, and operational decisions through a practical implementation approach.
Our initial engineering examples are predominantly electric. Gas and water examples are adaptations, with their own domain requirements. Read our editorial approach
WHAT GUIDES US
Practical progress, built on trust.
Start with useful work.
Find a real operational need, involve the people doing the work, and define an outcome they can recognize.
Build on what you have.
Connect existing systems and improve imperfect data as you go. A useful first step can be a small one.
Keep people accountable.
Make evidence, uncertainty, permissions, and review part of the workflow from the beginning.
Learn from the result.
Measure quality and the complete cost of the work. Expand when the evidence supports it, and share what you learn.
OUR SPONSOR
Supported by Senpilot.
AI for Utilities is an educational initiative sponsored by Senpilot, a company building AI software for utilities.
We make that relationship visible so readers can understand the perspective behind the guidance. Our aim is to make the learning useful across products, utility sizes, and implementation approaches.
Learn about SenpilotOUR APPROACH
Clear about our sources.
Our first edition combines a use-case catalog and anonymized utility experience with original implementation guidance. Here is how to read and use the material.
The use-case library (September 2026)
The library contains 139 named opportunities across engineering, regulatory, customer service, operations, and platform capabilities. These are catalog entries, not 139 validated deployments. Some describe enabling capabilities such as maps, exports, and recovery.
About value estimates
Catalogs like this one often carry directional value estimates with different assumptions, time horizons, and potential overlap. We do not publish or combine them as measured savings. Each entry describes potential value and how to evaluate it.
Utility implementation experience
Our guides and free downloadable resources draw on anonymized accounts of utilities building data foundations, running pilots, and putting AI into everyday work. These accounts are not independently audited studies. Worksheets, thresholds, and numerical examples are editorial and clearly labeled as illustrative.
Our editorial approach
Use-case titles, descriptions, and dataset requirements follow the catalog, with light editing for clarity. Suggested value, evaluation metrics, safeguards, and sector tags are editorial additions. Dataset lists are starting points, not complete integration specifications.
The engineering catalog is predominantly electric. Electric and Cross-utility are editorial tags added to each entry. Gas and water examples in articles are identified as adaptations requiring their own domain assessment.
Examples and calculator results illustrate a method, not a forecast for a particular utility. Build an investment case from your own baseline, operating costs, adoption, and output quality.
Guides include sources and review dates and link to primary references where useful, including the NIST AI Risk Management Framework. We distinguish source facts, editorial judgment, modeled benefits, and measured evidence.
Privacy in this edition
The site has no account registration, contact submission service, analytics scripts, or advertising trackers. Search matching and calculator calculations run in your browser. Calculator inputs are not sent to an application server.
Pages and the search index are requested from the hosting service, which may keep ordinary request logs. Submitted search terms can appear in the page URL. External services have their own privacy practices. This describes the current site implementation.