The first Actionable Intelligence tool for AI-driven storytelling

Any questions?

• Is my data being used to train the AI models

  No, your data is not retained on our servers beyond the duration of the interactive session.

  Zebra AI is a service running in the Microsoft Azure cloud. Your data is not retained on Microsoft Azure servers beyond the duration of the interactive session. We do not store it, it is firewalled from other users using Zebra AI, and we never use it for training, fine-tuning, or in any other way beyond what is needed to generate the answers that you see.

  The only external component used in generating your answers is OpenAI API which guarantees that neither inputs nor outputs are used for training or are in any way exposed to other accounts and retain the data for a maximum of 30 days. Only data we pass through to OpenAI is what has been asked in the interactive session, no user, personal, or broader context data are passed through. We try to minimize the amount of data sent to OpenAI by various means such as but not limited to doing inference and generation on our servers, substituting datasets with descriptive statistics, using sampling, …

• Security: what happens to my data once I upload it?

  Your data is completely secure when using Zebra AI. The data is uploaded to Zebra AI servers, hosted on Microsoft Azure, for the duration of the session. Once the session ends (e.g. the user signs out, refreshes or leaves the Zebra AI page), the uploaded data is immediately deleted from the server and Zebra BI does not story any information you provide. 

• What type of data can be uploaded? Are there any formatting requirements?

  In the first beta version of Zebra AI, we support single-table static data uploads. You can prepare your files as CSV or XLSX files. As with any other reporting tool, the data should be structured properly. Here are some practical tips.

  General:

  • Column header names & cell values should ideally be something meaningful (e.g., date, country, ID, etc.).

  CSV files:

  • Use comma-separated fields
  • For numerical fields the data should use the dot for the decimal separator and no thousands separator
  • Date or DateTime fields should use a standard format (ISO, yyyy-MM-dd HH:mm:ss, MM/dd/yyyy, etc.)

  Excel spreadsheets:

  • The first table on the first sheet will be read by the application
  • Your table should have headers only in the first row and data from the second row onwards
  • Empty values should be represented as empty cells
  • DateTime and numerical fields should be formatted in Excel accordingly.

• Is there a limitation to the size of the dataset I can upload (e.g. how many rows and columns)?

  While there is no hard limit on the number of rows, there is a limit to the uploaded file size, which is 50MB. When it comes to columns, up to 30 columns should not present difficulties to the system, but having noticeably more columns may result in errors due to GPT’s context window limitations. 

• Can I ask Zebra AI to generate a specific chart?

  Yes. In our beta version, you have the flexibility to select from 5 chart types: waterfall, pin, bar, integrated and line chart. Pin and bar chart can draw one numerical measure at a time, the waterfall and line chart may draw one or two measures and the integrated chart allows you to display up to two measures.