Relativity aiR - Legal AI Software

Relativity aiR offers three generative AI-powered software solutions built directly into RelativityOne: aiR for Review, aiR for Privilege, and aiR for Case Strategy. Together, they help legal teams power some of the most time-consuming parts of litigation and investigations.

Here are just a few ways legal teams use Relativity aiR:

• Document prioritization
• First-pass relevance and issue identification
• Conducting internal investigations
• Uncovering early case insights
• Privilege protection workflows
• Privilege log creation
• Fact extraction and timeline building
• Case strategy development

Relativity aiR is designed to be transparent, reviewable, and defensible, with safeguards inspired by Relativity’s AI Principles for responsible and useful AI. Key protections include:

• Document-level citations, rationales, and reasoning: aiR provides natural-language explanations backed by citations directly from the source document, so reviewers can immediately verify the model’s decisions and reasoning. This reduces hallucination risk and creates a clear audit trail.
• Human-in-the-loop, expert-led workflows: Subject matter experts define the instructions, guide the model, and validate outputs every step of the way. aiR amplifies and accelerates legal work but never replaces human judgement.
• Proven testing and validation: Our applied science team rigorously tests Relativity aiR models on real-matter data before deployment to ensure the technology meets the highest standards for quality and purpose-built performance. Once in use, aiR supports well-established processes for measuring recall, precision, and elusion metrics and performing continuous quality checks. These validation workflows help legal teams demonstrate reliability, consistency, and defensibility.

Relativity aiR is fully integrated into RelativityOne, which is built on Microsoft Azure and monitored 24/7/365 by Relativity’s in-house security team, Calder7. aiR uses the same enterprise-grade security protections used by Relativity’s global customer base, including legal teams at corporations, government agencies, and leading law firms.

• Customer data is always stored within the Relativity boundary.
• Customer data accessed by Relativity is extremely limited and done with a just-in-time access model.
• Relativity uses a privileged access management solution and classification schema that dictates how to handle confidential data.
• Relativity has opted out of any abuse and harmful content monitoring offered by Microsoft to ensure no unauthorized users have access to raw inputs or outputs.
• Relativity’s vendor risk management team performs an in-depth review of Azure’s security and privacy posture at least annually to validate their controls are operating effectively.

We’re always happy to talk security and answer any and all questions. Reach out to us for more information.

Relativity aiR for Review and aiR for Privilege are included in the standard pricing and packaging for RelativityOne. RelativityOne offers flexible pricing models to meet a legal team’s unique needs.

For more information – including pricing for aiR for Case Strategy – and to receive a tailored quote based on your legal team’s data volumes, workflows, and region, please contact our team.

No, you do not need be an AI expert. Relativity aiR is intuitively designed for legal professionals – not data scientists. What helps most are the skills you already use every day: giving clear instructions, thinking critically about evidence, and validating results.

Prompting in Relativity aiR is all about translating legal instructions into clear, structured guidance for the specific solution.

But if you do need help, Relativity provides extensive training, documentation, and hands-on guidance to help your team get started.

Relativity aiR is an agentic system that can eliminate repetitive, time-consuming cognitive tasks – producing better results, faster. Throughout its workflows, humans are in the loop: inputting and refining prompt criteria, validating outputs, evaluating the AI’s reasoning, and ensuring consistency.

We believe fully autonomous agents without human oversight do not belong in legal tech. It’s critical for legal professionals to guide, validate, and maintain control over AI’s decisions; human involvement ensures legal AI remains ethical, practical, and defensible.