Save time and improve accuracy by quickly transforming text queries into actionable visual insights, enhancing your decision-making process.
Ardoq's "Add new dataset with AI" functionality is powered by Microsoft’s Azure Open AI Service. Find out more about AI and innovation in Ardoq in this KB article.
We’d love your feedback! Share your thoughts on "Add new dataset with AI" by completing our brief survey.
How to Enable AI Functionality in Your Organization
To access “Add new dataset with AI”, "Refreshed Ardoq design and new Viewpoint builder" and "Add new dataset with AI" need to be enabled in your organization. You can check from the Preferences menu > Organization settings > Feature settings. These settings are available to admin users only.
If you don’t see the "AI Features" section in "Feature settings" available, contact your CSM to sign the updated Data Processing Agreement (DPA), which will unlock AI functionalities in Ardoq.
How to Automatically Create Viewpoints
To prompt the AI to create viewpoints:
Head over to the Viewpoints module on the left-side navigation bar
Open an existing viewpoint or create a new one
Once in the view, right-click on a component and select “Add new dataset with AI”
Type in your question and press enter. Alternatively, select one of the suggested queries.
See the new visualization that answers your query.
Follow the instructions on this KB article to save your viewpoint.
Please note: All users can create viewpoints with AI but only admin and writer users can save viewpoints.
FAQ
Is Ardoq partnering with any AI technology providers to offer this service?
Yes. Ardoq's AI features are powered by Microsoft’s Azure Open AI Service, which provides secure and efficient AI capabilities. Please find out more about this in the FAQ AI KB article.
Are my queries shared with third-party tools?
No. Users’ input data is not sent to any third party entities other than those authorized in your agreement. Find out more in the FAQ AI KB article.
Are my queries used to train other customers’ models?
No. Users’ input data is not used to train other customers’ models nor your own.
Where can I find more information about the language model, security and privacy?
Please refer to the FAQ AI KB article.
Is there a limit to how many questions I can ask at a time?
No, there is no limit. You can ask multiple questions simultaneously. However, please be mindful to clearly formulate each question to obtain accurate results.
What determines the accuracy of the graphs generated by the AI?
The accuracy of the graph depends on how clearly the questions are formulated, the available data in the organization, and the quality of that data. If no relevant data exists in the organization to answer a question, the AI may display a graph that only partially answers the query.
Version History
Version 1.0 (3 Jun 2024)
The current version of the prompt-based AI-generated traversals has some limitations:
It doesn't factor reference types.
It doesn't have good support for ordering of questions. For instance, "What servers do the applications which impact this initiative run on?" is a harder question for the AI to answer than "What servers does this application run on?"
It doesn't support general questions like "all references around this component".
These problems are actively being worked on. The best way to get good results with the current implementation is to think about asking for related component types, and the AI will seek the closest relationships to those component types.
As ever, please help us improve by completing our brief survey or by submitting feedback through our in-app chat for the prompts you'd like us to support!