Published on December 1, 2023, 7:36 am

Generative AI: LexisNexis Embraces Technology to Stay Ahead

In the fast-evolving world of legal information services, LexisNexis, a renowned industry giant, faces its most formidable challenger to date – generative AI. CTO Jeff Reihl acknowledges that generative AI is advancing at an unprecedented pace and has the potential to disrupt their business. To tackle this challenge head-on, LexisNexis is swiftly adopting and enhancing the technology.

LexisNexis has long been accustomed to adapting to competitive threats posed by evolving technologies such as the Internet, Google Search, and open-source software. However, generative AI represents a new level of disruption. Faced with this reality, the company’s leadership team held a meeting after OpenAI’s GPT-4 debuted in March last year. The consensus was clear – they needed to rewrite and reprioritize all their annual goals to address this game-changing innovation.

LexisNexis specializes in providing legal and business data and analytics services worldwide. This makes them particularly vulnerable to the impact of generative AI. Nevertheless, Reihl remains confident that his organization can overcome these challenges due to certain imperfections in current general-purpose large language models (LLMs) and their unique tools for enhancing and customizing LLMs used for their services.

To maximize value for their global customer base, LexisNexis has been working tirelessly with over 2,000 technologists and about 200 data scientists. They are incorporating unique features that leverage generative AI capabilities into their offerings. While this may seem like a recent development, LexisNexis has actually been exploring natural language processing (NLP) models such as BERT since Google introduced it in 2018.

The company now supports all major LLMs and utilizes both Microsoft Azure and AWS platforms when necessary. For instance, if customers engage with LexisNexis’ conversational search feature, their queries may be processed by both Azure’s ChatGPT-4 and AWS’s Anthropic simultaneously. LexisNexis strategically uses the optimal model to provide the most accurate answers to customer inquiries.

In a bid to address some of the shortcomings of existing LLMs, LexisNexis recently launched its own generative AI solution called Lexis+ AI. This innovative offering aims to eliminate AI “hallucinations” and provide linked legal citations, ensuring lawyers have access to up-to-date precedents. The company’s migration to the cloud, which began in 2015, has played a crucial role in enabling such advancements. While LexisNexis primarily operates on AWS, they also offer Microsoft Azure for customers using Microsoft Office and other related platforms.

The shift to the cloud was not without its challenges. When Reihl joined LexisNexis in 2007, approximately 50% of their infrastructure relied on mainframes. They operated two large data centers across the US and managed a diverse range of technologies and data formats due to various acquisitions. However, through strategic investments in XML-based open systems and subsequent rearchitecture for microservices in recent years, LexisNexis successfully migrated away from mainframes.

Apart from substantial cost savings, scalability, agility, and speed of innovation achieved through cloud migration, one of the most significant benefits for LexisNexis has been their ability to leverage machine learning and LLMs effectively.

The transformation within LexixNexix also required upskilling employees and acquiring new talent. The team now includes subject matter experts with legal backgrounds who understand legal language alongside UX designers, product managers, software engineers— plus nearly 200 data scientists and machine learning engineers.

With $1.4 billion invested in their digital transformation journey so far, LexisNexIs continues to drive innovation forward. The recent launch of Lexis+ AI in the US market features a multimodel LLM solution with generative AI enhancements, catering specifically to the legal industry. This AI SaaS platform offers advanced capabilities like retrieval augmented generative engine, refined conversational search, and document drafting, enabling users to analyze and extract key insights from legal documents within minutes.

The greatest challenge for LexisNexis, like many organizations, is finding enough skilled talent. Recognizing this obstacle, Reihl emphasizes the importance of training individuals with data acumen in these skills. With a talented team of 200 data scientists at their disposal, LexisNexis aims to expand and release their offerings in international markets over the next year.

In an ever-evolving landscape where generative AI poses new challenges, LexisNexis continues to embrace technology as a means to stay ahead. Through their innovative solutions and strategic investments, they are positioning themselves as leaders in leveraging the power of generative AI within the legal information services industry.

Paula Rooney is a senior writer at CIO.com

Share.

Comments are closed.