Release Notes - 2025.10.29

Welcome to the latest One Model product release update. This article provides an overview of the  product innovations and improvements to be delivered on 29 October 2025.

One AI 

• One AI Assistant Analyze - Analyze in One AI Assistant enables users to ask natural language questions about any chart or table in the assistant and get clear, contextual responses that simplify data interpretation. Analyze is available for both Visualizations and Answers and is powered by a secure large language model (LLM). The LLM is provided with both the user’s request along with the data displayed in the chart or table, enabling responses that are intelligent and grounded in the data. Think of it as securely pasting your data into ChatGPT and asking questions about it.
  
  Analyze is built directly into the assistant for in-flow analysis. Because conversations are threaded, users can ask follow-up questions that retain the full context of the discussion. Responses may include rich formatting such as lists, icons, and tables. As with all One AI Assistant capabilities, Analyze is permission-based and secure. Chart data is sent to the OpenAI LLM when using Analyze, so organizations should confirm alignment with their corporate AI policies prior to enabling this feature. 
  
  Please note that in this release the LLM leveraged by Analyze does not have the ability to search the web for current information. We plan to add optional web search capability in the 2025.12.10 release. Until then, requests for current information such as benchmarks are not available. (refs 3619 and 4038)
  
  
  
  Please see the One AI Assistant Analyze help article for more information.
  
   
• One AI Assistant Chat - Chat is a conversational AI feature built directly into One AI Assistant. It enables natural, back-and-forth dialogue with the assistant, whether you’re looking up HR concepts or brainstorming ideas. For example, you might ask about SHRM-recognized predictors of voluntary separation, get a concise answer, and then follow up with clarifying questions — all while staying in the same thread.
  
  Designed for ease of use, Chat requires no configuration. You simply switch to Chat mode and start asking questions. Responses can include formatting like lists, tables, and rich text that makes the information provided clear.
  
  Security and privacy are built in. No system data is sent to external services, and Chat operates in a permission-based environment with guardrails in place to prevent misuse. As with One AI Assistant Analyze, web search capability is not available to Chat in this release but is expected in the 2025.12.10 release. Until then, requests for current information such as benchmarks are not available. (refs 3736 and 4038)
  
  
  
  Please see the One AI Assistant Chat help article for more information.
   

Platform Updates

New Cookie Banner 

• To continue improving the One Model product, starting from this release (2025.10.29), we will be using a platform level analytics tool to collect and analyze information about how users interact with the product, helping us better understand performance, feature adoption and identify opportunities for enhancement.
   
• As part of this, we’ll be using analytics cookies. When you login you will notice a new notification banner where you will have the option to opt-in or opt-out of the use of these cookies along with links to our Cookies Policy and Privacy Policy. There will also be an option to change these settings in the user preferences page within One Model. These cookies are used solely to measure feature usage and performance within One Model, this is anonymous and we do not track user names or email addresses, or users across sites or serve advertising. 
   
• We wanted to introduce the notification banner for the highest level of transparency for users, but hope everyone does opt-in. Our goal is to use these insights to deliver a smoother, faster, and more valuable experience for you. Please reach out to us if you have any questions.

Data Improvements

• We've changed the Processing Script behaviour so that a user cannot modify the Processing Script of another user. This is designed to prevent unintended changes to a script someone else is working on that could cause Processing Script changes that are unwanted from making it into Production. (ref 20974)
   
• We now automatically generate the "daysinmonth" field for Time Dimensions in cases where it has not been configured. This field will soon be used to optimise the calculation for Rolling Month metrics. (ref 25667)
   
• When you delete a data source, a new checkbox is now available in the confirmation dialog. This allows you to choose whether to keep or remove the associated staging tables in your data warehouse.
  Please note: While this option is now visible in the user interface, the back-end functionality to fully support skipping table deletion is still in progress. As a result, the checkbox will not be fully functional until a future update. We will notify you as soon as this feature is live. 
  
  By default, the option to remove staging tables is selected, ensuring a clean and complete deletion. This change applies to all data source types, providing a more flexible and robust data management experience in the future. (ref 25133)
   
• You can now edit the schema name for a copied API connector data source before the initial data pull. After copying a data source, you'll be able to change the schema name.A warning will appear if you enter a schema name that is already in use for a data source of the same type, ensuring that all data source and schema combinations remain unique. (ref 16542)

Minor Improvements & Bugs Squashed

• Forecast Data Export - Forecasted data in Storyboard tiles can now be exported in CSV format. On any tile where a forecast has been run, simply click the lightbulb icon and select the Export Forecast CSV option. The file will automatically download but will not be listed on the Exports page. (ref 4066)
   

• We've improved the experience for selecting a timezone when configuring Storyboard Notifications.  Instead of manually scrolling through a long, time-consuming list of entries, users can now quickly locate the correct setting by typing the name of the timezone (e.g., "New York," "London," "Australia") directly into the input field. This change significantly streamlines the configuration process. (ref 25838)
   
• We made a small enhancement to drill-through on security restricted values. Previously, when a security restricted value was selected for drill-through in a Storyboard tile, the table showed values which could give the impression of circumventing security. This enhancement now blocks drill-through and shows a 'drill through not permitted' message. (ref 24451)
   
• We fixed an issue on the Site Validation page where certain error descriptions were incorrectly showing the placeholder tag {ownedByEmail} instead of the actual user's email address. (ref 25904)
   
• We fixed a bug in the Storyboard Define panel that affected drag-and-drop functionality for pivoted columns. Previously, when a large number of items were displayed, the drag-and-drop chips would sometimes appear out of order. (ref 25543)