Skip to main content

LexisNexis Unveils AI Assistant Protégé to Transform Legal Workflows

Legal research leader LexisNexis has introduced Protégé, a cutting-edge AI assistant tailored to support paralegals, legal assistants, and professionals in the legal industry. The tool is designed to enhance productivity by assisting with tasks such as drafting and proofreading legal documents, ensuring accurate citations in pleadings and briefs. Unlike generic AI tools, Protégé is built to learn the unique workflows of individual law firms, offering customized services that adapt to specific needs.

A Multi-Model Approach

In developing Protégé, LexisNexis experimented with advanced large language models (LLMs) from providers like Anthropic and Mistral. The company selected models based on their suitability for specific use cases, prioritizing performance and response times. In some instances, LexisNexis opted for smaller language models (SLMs) or employed model distillation, a technique where a larger model "teaches" a smaller one. This approach enhances efficiency and reduces costs while maintaining high accuracy.

Jeff Riehl, Chief Technology Officer at LexisNexis, explained the strategy: "We break down each user's task into individual components and find the best large language model to support each component." For example, when a user inputs a query, Protégé first uses a fine-tuned Mistral model to analyze the query before switching to the most appropriate model for task completion.

Building on a Legacy of AI Innovation

LexisNexis has a long history of integrating AI into its services, exploring technologies such as natural language processing, deep learning, and machine learning. The company established the LexisNexis + AI Legal Research Center in July 2024 to further its advancements in the field. Riehl highlighted the transformative impact of generative and conversational AI since the launch of ChatGPT in November 2022.

Expanding Capabilities

Currently, Protégé relies primarily on a fine-tuned Mistral model but plans to incorporate other models like Claude, OpenAI’s GPT series, and Google’s Gemini in the future. These additions aim to enhance the platform’s capabilities further.

Beyond Document Drafting

Protégé is more than just a document assistant. It helps generate questions for legal cases, link citations for accuracy, create timelines, and summarize complex legal materials. Riehl emphasized that Protégé represents an initial step toward personalized and intelligent legal assistance: "Our goal is to provide each legal professional with a personalized assistant that supports their specific work needs."

Key Points

  1. Protégé is an AI assistant designed to streamline legal workflows by drafting documents and generating citations.
  2. LexisNexis uses a combination of large and small language models to optimize performance and cost efficiency.
  3. The tool adapts to individual law firm workflows, offering customized support for legal professionals.
  4. Future updates will incorporate additional models like Claude and Google’s Gemini to enhance capabilities.

Enjoyed this article?

Subscribe to our newsletter for the latest AI news, product reviews, and project recommendations delivered to your inbox weekly.

Weekly digestFree foreverUnsubscribe anytime

Related Articles

Google's Project Genie Turns Sketches Into Playable Worlds
News

Google's Project Genie Turns Sketches Into Playable Worlds

Google DeepMind has unveiled Project Genie, an AI tool that transforms text prompts or simple drawings into interactive virtual environments. Combining multiple advanced AI models, it lets users instantly generate and explore personalized digital worlds—from alien landscapes to jungle adventures—with real-time responsiveness. Currently in early testing with strict time limits, this technology hints at revolutionary applications for gaming, filmmaking, and education.

February 2, 2026
AIVirtualRealityGoogleDeepMind
UBTech's Thinker Model: A Game-Changer for Smarter Robots
News

UBTech's Thinker Model: A Game-Changer for Smarter Robots

UBTech has open-sourced its Thinker model, a breakthrough in robotics AI that tackles critical challenges like spatial understanding and visual perception. By refining raw data from 20B to just 10M and slashing annotation costs by 99%, Thinker promises to revolutionize how robots learn and operate. This move could accelerate innovation across the robotics industry.

February 2, 2026
roboticsAImachine learning
Apple Faces Talent Drain as Key Researchers Join Meta and Google
News

Apple Faces Talent Drain as Key Researchers Join Meta and Google

As global competition in artificial intelligence intensifies, tech giant Apple is facing a severe talent retention challenge. After previously losing a senior Siri executive, at least four experienced AI researchers have recently left, joining major competitors such as AIbase Base

February 2, 2026
AIAppleMeta
News

Robotics Innovator LimX Dynamics Lands $200M Boost for Smart Machines

LimX Dynamics, a rising star in robotics intelligence, just secured a massive $200 million Series B funding round backed by global investors and industry leaders. The company is making waves with its groundbreaking TRON2 modular robot system and LimX COSA operating platform, which promise to revolutionize how robots interact with the physical world. Major players like JD.com and NIO Capital are betting big on this vision.

February 2, 2026
roboticsAIventure capital
OpenClaw: The Lobster AI That Finally Found Its Name
News

OpenClaw: The Lobster AI That Finally Found Its Name

The open-source AI assistant formerly known as Clawd has undergone its third rebranding, settling on OpenClaw after trademark hurdles and community feedback. Despite the naming drama, the project has exploded in popularity, surpassing 100,000 GitHub stars while maintaining its quirky lobster mascot. Offering local AI processing across multiple platforms, OpenClaw lets users manage emails, calendars and more while keeping all data private.

January 30, 2026
AIOpenSourcePrivacyTech
Ant Group's LingBot-VLA Brings Human-Like Precision to Robot Arms
News

Ant Group's LingBot-VLA Brings Human-Like Precision to Robot Arms

Ant Group has unveiled LingBot-VLA, a breakthrough AI model that gives robots remarkably human-like dexterity. Trained on 20,000 hours of real-world data, this system can control different robot arms with unprecedented coordination - whether stacking blocks or threading needles. What makes it special? The model combines visual understanding with spatial reasoning, outperforming competitors in complex tasks. And in a move that could accelerate robotics research, Ant Group is open-sourcing the complete toolkit.

January 30, 2026
roboticsAIAntGroup