Introduction
Nuclear vitality ranks among the many world’s most regulated industries. AI and particularly generative AI have created sufficient influence that thought leaders rank it amongst different transformative “common function applied sciences” comparable to electrical energy and the steam engine. Harnessing AI to reimagine nuclear operations throughout the trade means extra carbon-free nuclear vitality for electrical grids and information facilities, which the Worldwide Vitality Company estimates demand to double by 2026. In September 2024, Westinghouse unveiled its HiVE™ AI system, powered by its fine-tuned bertha™ generative AI mannequin, reworking how clients collaborate with Westinghouse.
Constructing a Higher Knowledge Administration Answer
Westinghouse’s digital transformation began greater than 5 years in the past with a deep bench of information and nuclear specialists and over 70 years’ price of cleaned and contextualized industrial information distinctive to the nuclear world. Nonetheless, the staff wanted to enhance the corporate’s information infrastructure if it wished to comprehend its AI ambitions. The prevailing on-premises analytics database lacked some necessary scalability options and choices. And not using a scalable cloud resolution, the info staff struggled with a scarcity of computing sources, an lack of ability to quickly experiment with large quantities of information, and restrictions on safely sharing information throughout functions.
To construct a world-class, nuclear-specific AI functionality, Westinghouse wanted a greater resolution. Westinghouse determined to construct on the Databricks Knowledge Intelligence Platform, a transfer that will show essential in its mission to drive innovation. The nuclear trade has all the time been deeply dedicated to security and lowering threat, with each element inspected and controlled. Managing and securing vital nuclear information will not be negotiable. With this in thoughts, Westinghouse got down to design an information spine that might host AI functions for among the most trusted utilities on this planet. Databricks was the best associate to assist Westinghouse obtain this objective.
As Westinghouse got down to design an information spine so safe and strong that it might host AI functions for among the most trusted utilities on this planet, it turned to Databricks. The Databricks staff rapidly grew to become a “guiding mild” for Westinghouse, offering essential assist because the Westinghouse infrastructure staff took the lead in configuring our methods to fulfill the nuclear trade’s strict regulatory necessities. Westinghouse was in a position to leverage Databricks’ state-of-the-art governance with Unity Catalog. It was constructed in accordance with finest practices outlined within the Databricks AI Safety Framework (DASF), complementing Microsoft’s strong safety requirements. These foundations bolster the credibility of Westinghouse’s information administration practices and provides its clients peace of thoughts, which is important in an trade the place belief and reliability are paramount.
When it got here time to modernize how the info was organized, the Databricks skilled providers staff delivered. Collectively, Westinghouse and Databricks created a scalable and multi-tiered analytics atmosphere, full with an ML Ops course of that streamlines the complete machine studying lifecycle. This basis additionally featured a sturdy prototyping atmosphere, together with devoted workspaces, for testing and deploying AI fashions, all backed by a safe and dependable information lakehouse structure.
The brand new infrastructure instantly saved tons of of hours yearly for the Digital Optimization Providers enterprise unit and allowed the Westinghouse staff to reinvest of their product strains to incorporate AI for customer-facing functions and providers.
To make this imaginative and prescient a actuality, Westinghouse had to make sure that its information was correctly ready, managed, and ruled. That’s the place Databricks’ highly effective applied sciences, together with Auto Loader, Photon engine, and Lakeflow Jobs, actually shined. Then, when Westinghouse wanted real-time insights into its information high quality and pipeline efficiency, they tapped into options like Lakehouse Monitoring and Expectations. Now, with Unity Catalog (UC) governing its information, Westinghouse has full visibility into its information’s journey, from supply to vacation spot. Within the nuclear trade, every part revolves round security and belief. As Westinghouse continues to develop pioneering new AI options, Databricks providers reinforce the belief Westinghouse earns for managing information securely and reliably.
Accelerating AI in a Complicated Trade
On September 4, 2024, Westinghouse launched its HiVE™ nuclear particular AI system and its bertha™ generative AI mannequin to the world. Not solely has the Westinghouse staff quickly superior its AI capabilities utilizing the Databricks Knowledge Intelligence Platform, however it may well now create future AI merchandise and options restricted solely by creativeness.
To help in growing bertha™, Westinghouse leveraged the Databricks Mosaic AI Agent Framework, to quickly consider numerous foundational fashions and GenAI methods. Utilizing Databricks Experiments and MLFlow, Westinghouse performed fast experimentation to find out the most effective fashions, whereas logging statistics to guage efficiency. This strategy enabled Westinghouse to speed up the event of its customized Generative AI resolution.
Westinghouse can now leverage its superior information infrastructure to create options throughout the nuclear trade. For instance, giant industrial amenities talk and retailer monumental portions of information. With an structure constructed on Databricks, Westinghouse maintains an AI resolution to extract, cleanse, and retailer machine information from over 200 nuclear amenities worldwide. One other instance consists of an AI utility designed to course of video information in real-time inside Strain Water and Boiling Water Reactors with the potential to detect particles not less than 90% higher than handbook inspections and save as much as 25% on inspection prices.
Lastly, one other nice instance consists of leveraging the bertha™ generative AI mannequin to generate licensing information and documentation dramatically sooner. Historically, it may well take months to manually compile new nuclear website licenses or environmental assessments. This can be a essential step in streamlining nuclear growth.
The Databricks infrastructure has freed information and nuclear specialists to give attention to nuclear innovation. Consequently, the Westinghouse information scientists delivered 4 proofs of idea in December 2024, two production-grade methods within the first quarter of 2025, and helped generate 45 distinct innovation concepts within the first two months of 2025.
Conclusion
The Westinghouse-configured Databricks Knowledge Intelligence Platform removes large limitations to attaining Westinghouse’s AI ambitions. Now, Westinghouse can scale compute, quickly and safely experiment with mass quantities of manufacturing information, and share data securely throughout functions. Westinghouse HiVE™ nuclear-specific AI system clients admire the ability of auditability, enter and output transparency, real-time information processing, and operational analytics. The Westinghouse groups worth the unimaginable and adaptable partnership with Databricks to create a novel platform that positions it for continued pioneering AI innovation.
“With Databricks all the time offering the newest options that hit the market, Westinghouse is ready to regularly incorporate new AI capabilities for our clients.”
— Catherine Stanley, Knowledge, Digital, and AI Supervisor at Westinghouse
