Last date modified: 2026-Jun-05

Developing prompt criteria

The prompt criteria are a set of inputs that provide aiR for Review with the necessary context it needs to understand the matter and evaluate each document based on the provided criteria. Developing these criteria is essential for training aiR for Review as a "reviewer," paralleling the process of preparing a human reviewer. See Best practices for tips and workflow suggestions.

Depending on the selected analysis type during set up, the prompt criteria panel in the dashboard will display different tabs. To assist in writing your initial prompt criteria, fields include italicized, grayed-out helper text offering example entries. Use it for reference when drafting your own criteria.

For certain analysis types, you may automatically generate initial prompt criteria from existing case documents, such as review protocols, requests for production or case memos, by using the prompt kickstarter feature. See Using prompt kickstarter for more information.

If needed, you can create project sets within a single aiR for Review project. This enables you to develop, validate, and implement prompt criteria without needing to create new projects for each iteration. See Using project sets for more information

To learn more about how prompt criteria versioning works and how versions affect the Viewer, see How prompt criteria versions affect the Viewer.

See these related resources for more information:

General prompt criteria workflow

We recommend the following workflow for developing and iterating on prompt criteria for all analysis types:

  1. For your first analysis, run the prompt criteria on a saved search of 50 test documents that are a mix of relevant and not relevant.
  2. Compare the results to human coding. In particular, look for documents that aiR for Review coded differently than the humans did and investigate possible reasons. This could include unclear instructions, needing to define an acronym or code word, or other blind spots in the prompt criteria.
  3. Tweak the prompt criteria to adjust for blind spots.
  4. Repeat steps 1 through 3 until aiR for Review produces results that reasonably align with coding decisions for the test documents.
  5. Test the prompt criteria on a sample of 50 more documents and compare results. Continue tweaking and adding documents until you are satisfied with the results for a diverse range of documents.
  6. Finally, run the prompt criteria on a larger set of documents.

aiR for Review only sees the extracted text of a document. It does not see any non-text elements, such as advanced formatting, embedded images, or videos. We do not recommend using aiR for Review on documents such as images, videos, or spreadsheets with heavy formulas. Instead, use it on documents whose extracted text accurately represents their content and meaning.

Entering prompt criteria

The tabs that appear on the prompt criteria panel depend on the analysis type you selected during set up. Refer to Setting up the project for more information.

  1. On the aiR for Review dashboard, you can enter data in the prompt criteria tabs using any of the methods below. Grayed-out, italicized helper text displays in the boxes as a guide for your entries.
    • Manually enter the required information as outlined in the Prompt criteria tabs and fields section.
    • Click the Draft with AI button (prompt kickstarter) to upload and use existing documentation, like requests for production or review protocols, to automatically fill in the tabs with draft criteria. This option only applies to Relevance, Key Documents, and Issues analysis types. See Using prompt kickstarter for more information.
    • Use project sets to develop, validate, and apply prompt criteria using a new data source (saved search) without having to create a new aiR project. See Using project sets for more information.
  2. Click Save after entering data on each tab or after all tabs are completed.
  3. After you have the desired prompt criteria set, click Analyze [#] documents to analyze documents. For more information on analyzing documents, see Running the analysis.
    After job analysis starts, tab fields cannot be edited, unless noted otherwise.
  4. To validate the prompt criteria, see aiR for Review prompt criteria validation and Setting up aiR for Review prompt criteria validation for more information.
    Performing prompt validation through aiR for Review and Review Center does not apply to Confidential Business Information analysis projects.
  5. If you want to use different saved searches within the project, see Creating a new project set.

Prompt criteria tabs and fields

Use the sections below to enter information in the necessary fields.

The combined total across all prompt criteria tabs must not exceed 15,000 characters.

Case Summary tab

The Case Summary gives the Large Language Model (LLM) the broad context surrounding a matter. It includes an overview of the matter, people and entities involved, and any jargon or terms that are needed to understand the document set. This tab only appears for Relevance, Key Documents, and Issues analysis types.

Relevance tab

This tab defines the fields and criteria used for determining if a document is relevant to the case. It appears if you selected Relevance or Relevance and Key Documents as the Analysis Type during project setup.

Key Documents tab

This tab defines the fields and criteria used for determining if a document is "hot" or key to the case. It appears if you selected Relevance and Key Documents as the Analysis Type during project setup.

Issues tab

This tab defines the issue fields and criteria used for determining whether a document relates to the set of topics or issues specified. It appears if you selected Issues as the Analysis Type during project setup.

Confidential Business Information tab

This tab defines the fields and criteria used for identifying Confidential Business Information (CBI) types and classifying documents by confidentiality level, such as Confidential, Highly Confidential, Attorneys' Eyes Only. It appears if you selected Confidential Business Information as the Analysis Type during project setup.

This functionality is being released through a phased rollout and is only available to clients in the US and EU Data Zone at this time. For more information on phased rollouts, see the Phased Rollout FAQ article available on the Community.

Using prompt kickstarter

aiR for Review's prompt kickstarter (Draft with AI option) lets you automatically generate initial prompt criteria for Relevance, Key Documents, and Issues analysis projects using existing case documents, such as review protocols, requests for production, complaints, or other foundational case documents, instead of doing so by scratch. By uploading up to 10 documents (with a total character count of up to 300,000), aiR for Review analyzes them to fill out the needed initial prompt criteria. This makes it easy to launch a new project with minimal effort. See Job capacity, size limitations, and speed for more information on document and prompt limits.

You can repeat this process as needed to refine the prompt criteria before starting the first job analysis. After the analysis begins, the Draft with AI option is disabled.

  • There are no additional charges to use prompt kickstarter.
  • Prompt kickstarter does not apply to Confidential Business Information projects.
  • To use prompt kickstarter for Issues projects, be sure to create an Issues field in the workspace beforehand starting the project.

Editing and collaboration

If two users edit the same prompt criteria version simultaneously, the most recent save will overwrite previous changes. For this reason, it can be beneficial to have only one user edit a project's prompt criteria at any given time. Defining distinct roles for users when updating prompt criteria may help streamline the process.

To facilitate team collaboration outside of RelativityOne, use the Export option. It exports the contents of the currently displayed prompt criteria to an MS Word file. For more information, see Exporting prompt criteria.

How prompt criteria versioning works

Each aiR for Review project comes with automatic versioning controls so that you can compare results from running different versions of the prompt criteria. Each analysis job that uses a unique set of prompt criteria counts as a new version.

When you run aiR for Review analysis, the initial prompt criteria are saved as Version 1. Edits to the criteria create Version 2, which you can repeatedly modify until you finalize by running the analysis again to see the results. Subsequent edits follow the same pattern, creating new versions that finalize with each analysis run.

To see dashboard results from a earlier version, click the down arrow next to the version name in the project details strip. From there, select the version you want to see.

Version selector

How prompt criteria versions affect the Viewer

When you select a prompt criteria version from the dashboard, this also changes the version results you see when you click on individual documents from the dashboard. For example, if you are viewing results from Version 2, clicking on the Control Number for a document brings you to the Viewer with the results and citations from Version 2. If you select Version 1 on the dashboard, clicking the Control Number for that document brings you to the Viewer with results and citations from Version 1.

When you access the Viewer from other parts of Relativity, it defaults to showing the aiR for Review results from the most recent version of the prompt criteria. However, you can change which results appear by using the linking controls on the aiR for Review Jobs tab. For more information, see Managing aiR for Review jobs.

Return to top of the page
Feedback