SAS Visual Investigator Copilot - Assisted Search

SAS Viya Copilot is a set of SaaS features that utilizes LLMs to provide end users with a more intuitive and accessible way to work within the SAS Viya ecosystem. This post is part one of a series exploring SAS Viya Copilot for Visual Investigator, with an introduction to SAS Viya Copilot and a deep dive into its assisted search feature for Visual Investigator.

SAS Viya Copilot

SAS Viya Copilot is available to those who are on SAS Viya 4, with a release version of 2025.12 or later, on an active internet connection. The Copilot features for Visual Investigator was first released for public preview in version 2026.02 and became generally available in 2026.04.

A few key points to mention regarding SAS Viya Copilot:

No customer data is used in the training of the Large Language Models
SAS Viya securely forwards requests to the SAS Viya Copilot service in SAS’s Microsoft Azure environment specific to SAS, which then issues one or more prompts to an LLM hosted there.
Customer data are not retained by SAS or Microsoft

It’s important to note that it’s crucial you review any output generated by AI to ensure its accuracy.

AI can and do make mistakes.

In order to use Copilot, it must first be activated in the SAS Viya Copilot page in SAS Environment Manager (Mange Environments). Prior to activation, ensure that you’ve set up the appropriate credentials and secrets for Copilot.

Select any image to see a larger version.
Mobile users: To view the images, select the "Full" version at the bottom of the page.

SAS Viya Copilot for Visual Investigator

There are several ways to use Copilot to enhance your SAS Visual Investigator usage. You can use the SAS Viya Copilot pane to:

Learn how to perform a task in SAS Visual Investigator
- Example: Ask Copilot how to perform a map search or how to disposition an alert and it will give you a step-by-step how-to.
Better understand the configuration of your particular SAS Visual Investigator implementation
- Example: Ask Copilot which object types have been configured by your administrator, and it will list out the exact object types, giving you an insight into your system configuration.

In addition to surfacing insights about your SAS Visual Investigator instance through conversational chat within the Copilot pane, there are three in-context experiences that have been directly embedded:

Assisted search: Use natural, conversational language to search in SAS Visual Investigator.
Image description generation: Use GenAI to create descriptions for images that appear on a page.
Prompt-based text generation: Use GenAI to create text based on a preconfigured prompt.

Feature Enablement

Depending on your software version, the abovementioned features may first need to be toggled on within SAS Environment Manager. Note that this is only relevant for software releases older than 2026.04, as the features become generally available in 2026.04 and onwards.

The relevant features are:

GenAI Assisted Search
GenAI Image Description
GenAI Text Generation
GenAI Conversational Agent

Capabilities

Additionally, in order for a user to perform an assisted search, they must belong to a group with the “Perform searches assisted by generative AI” capability (svi.search.assisted).

Using Assisted Search

The assisted search feature will take the user’s natural language query and automatically translate it into the appropriate context within SAS Visual Investigator, as well as displaying the returned results using the most appropriate view.

I will cover three examples of how to use the assisted search feature in this post.

Example 1: Text Search

To conduct a simple assisted text search, navigate to the Search page, set the Type to Standard search, and select the Assisted search radio button.

In the example below, I’ve searched for “Alerts with scores higher than 250, displayed in a table”. Notice that I’ve specified the object type (Alerts), a condition on the object attribute (alert scores > 250), as well as how I would like to view the data (in a table format).

See below for the returned results, where alerts with scores greater than 250 are returned as a detailed table.

Example 2: Relationship Search

In the second example, I’m interested in finding vehicles that are linked to a particular auto policy through all the auto claims that have been filed against that policy. This is a complicated search, leveraging the linkage between multiple object types. This would typically be done through a relationship search. You can learn more about relationship search in my previous post Relationship Search in SAS Visual Investigator.

To leverage assisted search to perform this search, I’ve simply entered the search prompt: “Find vehicles linked to claims to policy 35432”.

Copilot correctly deduces that my query is best suited for a relationship search and asks me whether I’d like to proceed forward with a relationship search or continue with a standard search.

The search result is shown below, where my result is automatically displayed in a network, showcasing all the interconnections between the different vehicles, the linked claims, and the target auto policy.

Example 3: Map Search

Lastly, let’s look at how Copilot handles a search involving geographical data. Starting on the same standard search page with the Assisted search radio button selected, I typed in the search prompt “I want to see home claims that happened on Pembury Road”.

Again, Copilot has correctly deduced the best type of search for my search prompt—map search.

The result is shown below, in map view. As you can see, it was able to correctly find the home claim whose address is listed as Pembury Road.

Assisted search is a fantastic feature that would significantly enhance the user’s search. Being able to use natural, conversational language to search within the data repository is a true game changer.

Summary

In this post, I provided an introduction to SAS Viya Copilot for Visual Investigator and a deeper dive into the assisted search feature. In the next post, I will cover the remaining features:

Image description generation
Prompt-based text generation

I will provide detailed examples on how to use these features in the SAS Visual Investigator’s Investigator UI (Investigate and Search Data), as well as the necessary configuration steps in the Admin UI (Manage Investigation and Search).

Stay tuned!