March 11, 2025
In the past few years, generative AI has found applications in virtually every industry, helping business streamline tasks to save time and reduce costs. For digital twins, generative AI can support digital twin development, data analysis, and intuitive data exploration. These benefits make it easier to manage digital twins than ever before.
ScaleOut Digital Twins™ is proud to announce the integration of digital twins and generative AI with a set of new AI features in our user interface. ScaleOut Digital Twins' new AI Assistant adds the analytical power of generative AI to our digital twin platform. Now, you can ask questions and get answers, generate queries and visualizations, and automatically monitor data for trends and anomalies.
The new generative AI features are designed to complement and enhance the existing features of the ScaleOut Digital Twins UI. The UI already enables users to create queries, which select a filtered set of digital twin instance data and display them on a map or table, and widgets, chart visualizations that monitor live digital twin data. By adding generative AI, queries and widgets don't just aggregate and display data--they can also provide AI-powered analysis. In addition, users can now create queries and widgets with just a brief text description, making this process simpler and more intuitive.
ScaleOut Digital Twins now lets users add AI-powered anomaly detection to any query or widget with just a few clicks. When creating a new query or widget in the ScaleOut Digital Twins UI, just check the box for "AI-Powered Trend and Anomaly Detection." The AI Assistant will then begin gathering and monitoring data, sending alerts if it finds any issues in the aggregated data.
ScaleOut Digital Twins gives you control over anomaly detection to suit your needs. For example, you can adjust:
For very fast-changing data, you might want to check for anomalies every minute, while other cases might only require checks every half hour. Whatever your preference, the AI Assistant acts as an extra pair of eyes that reviews your data without the need for manual oversight.
If the ScaleOut Digital Twins AI Assistant finds any notable trends or anomalies in the data, it sends an alert message. The message uses generative AI to summarize the issue, describing when it occurred and what happened. The description also includes links to the relevant model, query, or widget, so you can quickly investigate the issue. A list of all AI-generated notifications can be found in the new Insights page of the ScaleOut Digital Twins UI.
With new generative AI features, the ScaleOut Digital Twins UI lets you create queries and charts with just a description. Just type in what you'd like to create, and the AI Assistant will build it automatically.
To generate a query or widget with AI, switch to the "Generate" tab in the UI. Then, type in a description of the query or widget you'd like to make, or ask a question that can be answered with a query or widget. For example, imagine you are an operational manager monitoring a network of sensors. You could ask a question like "How many sensors are reporting signal strength below 2 and latencies greater than 10 ms?" The AI assistant then translates this text description into a query. Complete sentences and perfect grammar aren’t needed; an even simpler request, such as "signal strength < 2 and latency > 10" can achieve the same result.
If you have a question about your digital twin model but you're not sure whether a query or widget is best, no problem. To ask the AI Assistant an open-ended question, go to the model's page and open the AI Assistant tab. Type in your question, and the Assistant will suggest a query or widget that it can create. The Assistant will also let you know if the request is beyond the current abilities of the system and will suggest alternatives.
To add generative AI to our digital twin platform, ScaleOut Digital Twins connects to the OpenAI API. When you talk to the AI Assistant, or when we scan your digital twins' data for anomalies, the aggregated data is sent to the API along with a system prompt. We then process the output data from OpenAI and display the results in the UI.
The ScaleOut Digital Twins integration of generative AI is implemented with performance in mind. OpenAI API requests are made asynchronously in the background, so they don't interrupt the flow of anything else you may be doing in the UI. Data is sent to OpenAI in an optimized format to minimize the number of tokens used. And if you'd like to temporarily disable AI features, you can do so at any time by simply checking a box in the Administrator settings.
ScaleOut knows that data accuracy is a critical concern when using generative AI systems for digital twins. We take steps to prevent AI hallucinations, which occur when a generative AI gives incorrect or outdated information in its responses. To avoid AI hallucinations, we use structured data outputs and constrain responses to ensure AI-generated content is factually based on your digital twin data.
ScaleOut Digital Twins' integration of digital twins and generative AI streamlines data management and analysis. Its capabilities help reduce the workload for operational managers. A set of exciting new features enables users to generate queries and visualizations with text prompts, easily use AI-powered anomaly detection, and explore AI-generated content at a glance. Our AI features are designed to integrate seamlessly into the user experience and provide accurate, informative results.
Devon Bain, UX Engineer at ScaleOut Software
Devon joined ScaleOut Software in 2023. He has contributed extensively to the design and development of the ScaleOut Digital Twins UI, including the integration of AI-powered features. He holds an MS in Information Systems from Cornell University with a focus on connective media.
.jpg)
