Everlaw Debuts New GenAI Feature, Allowing Legal Teams to Search Terabytes of Ediscovery at Once - and in Seconds

Project Query delivers game-changing ability to surface critical insights for litigation and investigative work

OAKLAND, Calif.--(BUSINESS WIRE)--Everlaw, the cloud-native investigation and litigation platform, today announced a new GenAI-powered addition to the Everlaw^AI Assistant portfolio, code-named Project Query, available in a closed beta program and to be demoed at Legalweek 2025, delivering an AI breakthrough for legal teams to expedite the investigation of a large corpus of documents in seconds.

Finding the right evidence is the cornerstone of resolving legal cases. The advent of the GenAI technique called retrieval-augmented generation (RAG) promised a long sought-after capability to retrieve information exclusively within a large corpus of ediscovery in natural language. Yet the application of RAG in legal matters is stymied by the complex nature of those docs. Unlike knowledge bases or technical documentation, legal discovery, such as emails, texts and chats, are often unstructured, nuanced and require understanding context across multiple domains. In addition, the kinds of questions legal teams pose will change as the case evolves.

How Project Query Works: RAG, Refinement and Reasoning

Everlaw^AI Assistant’s Project Query aims to deliver on the promise of RAG, allowing legal teams to ask questions across terabytes of ediscovery at once, while compensating for legal’s ‘unfriendly’ corpus and ambiguous exploration with an innovative process of refinement to draw relevant facts, insights and key documents with greater accuracy and reliability. Everlaw designed Project Query with the idea that legal teams may ask about more nuanced elements: behaviors, intent, actors and conditions. Query has a sophisticated refinement process to find relevant evidence for natural language queries.

Everlaw has paired its refinement approach with the latest in LLM reasoning models, which make more logical inferences and draw more nuanced conclusions that are highly valuable to legal cases. In essence, these models mimic how humans think through a problem by breaking it down into smaller, logical steps.

With those combined capabilities, Project Query is designed to uncover insights in an entire corpus of data sooner by simply asking questions related to specific issues, parties, or events and get answers in everyday conversational language in just seconds. Answers are supported with a list of facts and referenceable resources so users can dive deeper into the breakdown of information available as part of the response.

Comments on and Responses to Project Query

“Pinpointing facts in a vast corpus is gold and doing it in seconds is game-changing,” said Steven Delaney, Litigation Support Director, Benesch. “Project Query feels more elegant than keyword search, delivering answers—not just documents—instantly. By getting straight to the facts, we can save time, prioritize strategy and reshape the discovery workflow.”

“Everlaw continues to raise the bar in GenAI innovation with Project Query,” said Ryan O’Leary, research director at IDC. “It’s one of the most practical AI applications I’ve seen in discovery—enabling investigation before review in seconds. With its ease of use, Project Query promises to instantly boost efficiency and deliver a significant impact for legal teams.”

“Project Query holds great promise for legal professionals – especially senior attorneys who for the first time can interrogate massive document sets at once, applying their honed expertise and instincts to find key evidence,” said AJ Shankar, founder and CEO of Everlaw. “In the AI arms race now shaping up in courtrooms across the country, Project Query could give tech-savvy legal teams an unmistakable edge.”

To give legal professionals confidence in the results of their Project Query work, Everlaw designed it to ensure:

• Answers are solely based on what’s in the document set
• Answers are verifiable with referenceable sources
• No client data is stored or retained by the LLM

Availability

In this closed beta testing of Project Query, Everlaw will provide access to a select number of customers to help test and iterate on the product. Everlaw conducted a year-long beta for its AI Assistant, which was generally available in Fall 2024. For Project Query, Everlaw anticipates general availability this year.

About Everlaw

Everlaw helps legal teams navigate the increasingly complex ediscovery landscape to chart a straighter path to the truth. Trusted by Fortune 100 corporate counsel, 91 of the Am Law 200, and all state attorneys general, Everlaw's combination of intuitive experience, advanced technology, and partnership with customers empowers organizations to tackle the most pressing technological challenges—and transform their approach to discovery and litigation in the process. Founded in 2010 and based in Oakland, Calif., Everlaw is funded by top-tier investors, including Andreessen Horowitz, CapitalG, HIG Growth Partners, K9 Ventures, Menlo Ventures, and TPG Growth. Follow us on LinkedIn.