Be a part of our day by day and weekly newsletters for the newest updates and unique content material on industry-leading AI protection. Be taught Extra
When authorized analysis firm LexisNexis created its AI assistant Protégé, it needed to determine one of the simplest ways to leverage its experience with out deploying a big mannequin.
Protégé goals to assist legal professionals, associates and paralegals write and proof authorized paperwork and be certain that something they cite in complaints and briefs is correct. Nevertheless, LexisNexis didn’t need a normal authorized AI assistant; they needed to construct one which learns a agency’s workflow and is extra customizable.
LexisNexis noticed the chance to carry the facility of huge language fashions (LLMs) from Anthropic and Mistral and discover one of the best fashions that reply consumer questions one of the best, Jeff Riehl, CTO of LexisNexis Authorized and Skilled, instructed VentureBeat.
“We use one of the best mannequin for the particular use case as a part of our multi-model method. We use the mannequin that gives one of the best consequence with the quickest response time,” Riehl mentioned. “For some use instances, that can be a small language mannequin like Mistral or we carry out distillation to enhance efficiency and scale back value.”
Whereas LLMs nonetheless present worth in constructing AI purposes, some organizations flip to utilizing small language fashions (SLMs) or distilling LLMs to turn out to be small variations of the identical mannequin.
Distillation, the place an LLM “teaches” a smaller mannequin, has turn out to be a well-liked technique for a lot of organizations.
Small fashions typically work finest for apps like chatbots or easy code completion, which is what LexisNexis needed to make use of for Protégé.
This isn’t the primary time LexisNexis constructed AI purposes, even earlier than launching its authorized analysis hub LexisNexis + AI in July 2024.
“We have now used lots of AI previously, which was extra round pure language processing, some deep studying and machine studying,” Riehl mentioned. “That actually modified in November 2022 when ChatGPT was launched, as a result of previous to that, lots of the AI capabilities have been type of behind the scenes. However as soon as ChatGPT got here out, the generative capabilities, the conversational capabilities of it was very, very intriguing to us.”
Small, fine-tuned fashions and mannequin routing
Riehl mentioned LexisNexis makes use of completely different fashions from many of the main mannequin suppliers when constructing its AI platforms. LexisNexis + AI used Claude fashions from Anthropic, OpenAI’s GPT fashions and a mannequin from Mistral.
This multimodal method helped break down every job customers needed to carry out on the platform. To do that, LexisNexis needed to architect its platform to swap between fashions.
“We might break down no matter job was being carried out into particular person elements, after which we’d establish one of the best massive language mannequin to assist that element. One instance of that’s we are going to use Mistral to evaluate the question that the consumer entered in,” Riehl mentioned.
For Protégé, the corporate needed sooner response occasions and fashions extra fine-tuned for authorized use instances. So it turned to what Riehl calls “fine-tuned” variations of fashions, primarily smaller weight variations of LLMs or distilled fashions.
“You don’t want GPT-4o to do the evaluation of a question, so we use it for extra refined work, and we swap fashions out,” he mentioned.
When a consumer asks Protégé a query a few particular case, the primary mannequin it pings is a fine-tuned Mistral “for assessing the question, then figuring out what the aim and intent of that question is” earlier than switching to the mannequin finest suited to finish the duty. Riehl mentioned the subsequent mannequin could possibly be an LLM that generates new queries for the search engine or one other mannequin that summarizes outcomes.
Proper now, LexisNexis largely depends on a fine-tuned Mistral mannequin although Riehl mentioned it used a fine-tuned model of Claude “when it first got here out; we’re not utilizing it within the product as we speak however in different methods.” LexisNexis can be fascinated about utilizing different OpenAI fashions particularly because the firm got here out with new reinforcement fine-tuning capabilities final yr. LexisNexis is within the means of evaluating OpenAI’s reasoning fashions together with o3 for its platforms.
Riehl added that it could additionally have a look at utilizing Gemini fashions from Google.
LexisNexis backs all of its AI platforms with its personal data graph to carry out retrieval augmented era (RAG) capabilities, particularly as Protégé may assist launch agentic processes later.
The AI authorized suite
Even earlier than the appearance of generative AI, LexisNexis examined the potential for placing chatbots to work within the authorized {industry}. In 2017, the firm examined an AI assistant that might compete with IBM’s Watson-powered Ross and Protégé sits within the firm’s LexisNexis + AI platform, which brings collectively the AI companies of LexisNexis.
Protégé helps legislation companies with duties that paralegals or associates are inclined to do. It helps write authorized briefs and complaints which might be grounded in companies’ paperwork and knowledge, counsel authorized workflow subsequent steps, counsel new prompts to refine searches, draft questions for depositions and discovery, hyperlink quotes in filings for accuracy, generate timelines and, in fact, summarize advanced authorized paperwork.
“We see Protégé because the preliminary step in personalization and agentic capabilities,” Riehl mentioned. “Take into consideration the various kinds of legal professionals: M&A, litigators, actual property. It’s going to proceed to get an increasing number of personalised based mostly on the particular job you do. Our imaginative and prescient is that each authorized skilled may have a private assistant to assist them do their job based mostly on what they do, not what different legal professionals do.”
Protégé now competes towards different authorized analysis and expertise platforms. Thomson Reuters personalized OpenAI’s o1-mini-model for its CoCounsel authorized assistant. Harvey, which raised $300 million from buyers together with LexisNexis, additionally has a authorized AI assistant.
