AI products and features in RelativityOne

This topic provides a high-level overview of aiR and other AI features in RelativityOne.

aiR suite

Relativity’s aiR suite is a set of AI capabilities including generative AI built into RelativityOne to accelerate and enhance legal data intelligence work.

• aiR for Review uses generative AI and natural language processing to analyze extracted document text and images based on prompt criteria. It helps pinpoint significant documents, explains their significance, and provides citations from those documents.
• aiR for Privilege uses generative AI, natural language processing, and machine learning to analyze documents and identify relevant personas, helping to predict whether documents may be privileged, and to generate a draft privilege log description.
• aiR for Case Strategy uses generative AI to assist in building your case strategy. It supports robust fact extraction, timeline development, deep summarization capabilities, and work product creation.
• aiR for Data Breach Response uses generative AI to support and help streamline the process of identifying, extracting, and linking Personal Information (PI) and Personal Health Information (PHI) to the relevant individual or organization, and generates a consolidated list of potentially impacted individuals.
• aiR Assist is a generative AI conversational search tool integrated within RelativityOne, designed to empower legal teams to interact with their data using natural language.

aiR systems explained

Relativity provides aiR systems that power applications and features across the RelativityOne platform. An aiR system takes a set of inputs and uses artificial intelligence technology to generate output. Inputs can include information/instructions provided by the user, information/instructions provided by Relativity, and content from documents, fields, and other information stored in Relativity. The AI technology behind a system can be a machine learning model (such as a classifier), a generative AI model, a third-party AI service, other AI technologies, or a combination of these. Think of an aiR system as a recipe: it specifies the ingredients and how they work together to produce a result.

Data Security

Customer data processed using aiR systems is not used to train or improve third party models or services. For more information regarding Relativity’s privacy and security for aiR systems, please see the AI security whitepaper.

Deployment, inference, and processing geographies

Relativity aiR systems currently use AI models and conduct AI processing using Microsoft Azure services unless otherwise specified in the terms of your Agreement with Relativity. When the data you select is processed by an aiR System (a process also known as AI inference), such processing may take place outside the Microsoft Azure region where your RelativityOne instance is hosted. Processing regions vary by product and capability and are determined based on AI model availability that provides the latest models available with the scale and performance required for each aiR system to meet Relativity’s minimum product requirements. Your data remains stored in the region you selected, only the aiR system AI inferencing may occur elsewhere. For clarity, when this happens your data continues to be protected in accordance with the terms of your Agreement (in particular, the data security terms incorporated into your Agreement).

RelativityOne instance regions	aiR system inference location
United States	United States
France, Germany, Ireland, Netherlands	Germany* and EU Data Boundary
United Kingdom, United Arab Emirates	United Kingdom
Switzerland	Switzerland
Brazil, South Africa	EU Data Boundary
Australia	Australia
Canada	Canada
Hong Kong, Japan, Singapore, South Korea	Japan
India	India
*AI processing in Germany will move to the EU Data Boundary on October 1, 2026. For more information about the EU Data Boundary, see this Microsoft article: What is the EU Data Boundary?

For RelativityOne Government customers, RelativityOne Government leverages the Azure Government Cloud with data centers located within the continental United States. All customer data in RelativityOne Government is processed in the continental United States.

Large Language Models (LLMs) used by aiR systems

Relativity’s aiR systems currently leverage a combination of generative AI LLMs including GPT-4o, GPT-4.1 mini, GPT-5 mini, GPT-5.1, and GPT-5.2 through Microsoft Azure services. The specific LLM or combination of LLMs used by an aiR system depends on the aiR system objective, and the capability and availability of an LLM that meets Relativity's minimum product requirements.

AI rate-limiting

AI-driven features in RelativityOne, including aiR capabilities, depend on shared platform resources and third-party large language model (LLM) providers with finite capacity. To protect service performance, manage demand, and support equitable access across users and instances, Relativity may, in its sole discretion, rate-limit, queue, prioritize, or restrict usage of these features. During such periods, you may experience longer wait times, queued jobs, reduced throughput, or temporary submission limits. Relativity determines and applies these platform-level controls independently of any job size limits your administrators configure at the instance level. Relativity may adjust these controls from time to time.

aiR Units and billing

Some aiR products and features may consume aiR Units or incur fees for use of aiR products as explained in a customer’s contract. See the topic aiR Units in RelativityOne standard workspaces for information on aiR Units.

aiR system updates

Relativity regularly updates aiR systems by modifying Relativity inputs/instructions, AI technology, and other features to improve the outputs of the aiR systems. The goal is to improve quality, accuracy, consistency, or responsiveness depending on the product and task. As a result, customers may notice changes in the product experience or differences in outputs after an update, even when the same prompt criteria or workflow is used. As with other Relativity software updates, we will communicate changes to aiR systems through product release notes, knowledge base articles, and/or official announcements.

Other AI-integrated features

In addition to the aiR tools listed above, other products and features in RelativityOne also utilize or integrate with AI. These include:

• AV Transcription analyzes audio and video files, then uses AI to transcribe spoken words into text. This streamlines the process of reviewing AV files and makes it easier to search for keywords, find correlations, and locate relevant files.
• Conceptual Analytics uses Latent Semantic Indexing (an early form of AI) to discover relationships between documents, even if they don't share the same key terms and phrases.
• Contracts extracts data points from contracts into Relativity fields using generative AI, custom ML models, or regular expressions.
• RelativityOne FOIA is a modular, AI-powered ecosystem that streamlines intake, tracking, review, publication, and reporting for FOIA requests.
• RelativityOne MCP provides an interface that allows AI assistants access to RelativityOne capabilities through Model Context Protocol, an open standard for connecting AI tools to external data and services through MCP clients.
• Prioritized Review in Review Center uses an AI classifier to learn from reviewers' coding decisions and prioritize relevant documents. This can be combined with aiR for Review to help boost review speeds.
• Sentiment Analysis uses AI to score documents according to the likelihood that they contain certain emotions.
• Translate utilizes machine translation to translate documents in different languages to another language, in support of searching and review workflows.