Relativity introduces conversational AI and no-code custom analyses for legal data

Relativity added aiR Assist and custom analyses to aiR for Review, indexing up to 300,000 documents per index. Legal teams can query data with plain language and define workflows.

Categorized in: AI News Legal
Published on: Jun 16, 2026
Relativity introduces conversational AI and no-code custom analyses for legal data

Relativity announced the general availability of aiR Assist and custom analyses in Relativity aiR for Review at its RelFest London event on June 16, 2026. Both capabilities will be standard in RelativityOne by the end of June, giving legal teams early case intelligence through plain-language questions and the ability to define their own AI-driven document review workflows without writing code.

aiR Assist lets legal professionals ask questions of their data as soon as it enters RelativityOne and receive citation-backed answers instantly. The system indexes up to 300,000 documents per index and supports workspaces with 1.5 million documents, allowing teams to find facts, trace evidence, and shape strategy from the start of a matter.

"Almost any AI solution can read a document. Almost none can be the single, auditable source of truth for every piece of data in your matter," said Phil Saunders, CEO of Relativity. "That's the difference: insights grounded in actual data, outputs defensible by design, and governance that holds up when it counts, in court, before a regulator or in front of a board."

Early insight without leaving the platform

When international advisory firm Interpath needed to identify critical evidence from over 30,000 unstructured documents in a multi-jurisdictional fraud investigation, investigators used aiR Assist to ask plain-language questions and follow investigative threads. The tool surfaced an email from years earlier that directly supported their case.

"With aiR Assist, we can ask natural questions of the data and surface critical evidence and actionable answers in minutes," said Gina Bieber, Director at Interpath. "It is the easiest tool to use in the history of Relativity. You just ask it a question, and it answers."

Because answers are grounded in data already governed within RelativityOne, every output remains traceable and explainable. Nikkisha Z. Scott, Consulting Attorney for Discovery & Litigation Technology at Cleary Gottlieb, described the tool as "a force multiplier" that lets lawyers work through discovery conversationally, surface insights faster, and deliver stronger results.

As AI for Legal continues to reshape how legal teams operate, Relativity is embedding these capabilities directly into the platform. aiR Assist will also power interactive memo drafting in Relativity aiR for Case Strategy, included in RelativityOne starting July 1.

Custom analyses that match the matter

Custom analyses extend aiR for Review's existing classification tools. Teams describe what they need in natural language-document classification, data extraction, or image analysis-and apply the analysis across an entire document set. An iterative prompt experience lets users test and validate outputs before running at scale.

"One legal team may need to identify key information from medical records, another may need to classify workplace communications, another may be looking for evidence of construction defects," said Andrew Milauskas, Chief Operating Officer at Page One, Inc. "Through natural language prompts, teams can easily customise analyses tailored to their specific objectives and get to meaningful results almost immediately."

One platform, multiple work surfaces

Relativity also previewed how it plans to bring AI-native drafting into Microsoft Word through its acquisition of Gavel. The vision is that aiR-generated content could open directly in Word, where lawyers draft, redline, and finalise, with edits syncing back to RelativityOne in real time. This keeps more legal work connected to the underlying data.

The company's innovation arm, Rel Labs, is backing tools for courtroom simulation, IP litigation, and case strategy. Additionally, a Model Context Protocol integration with Claude allows legal data and technology teams to handle administrative tasks-such as setting up matters and managing access-through plain-language prompts, while the substantive legal work stays governed within RelativityOne.

Why this matters for legal professionals

These updates reduce the lag between data intake and actionable insight. Teams no longer need to toggle between separate tools or rely on predefined review categories. Instead, they can interrogate their data conversationally, define custom document analyses in seconds, and trust that every output ties back to an auditable record-a critical requirement when work faces scrutiny in court or before regulators.


Get Daily AI News

Your membership also unlocks:

700+ AI Courses
700+ Certifications
Personalized AI Learning Plan
6500+ AI Tools (no Ads)
Daily AI News by job industry (no Ads)