The Growing Impact of Generative AI on the Legal Industry: What Lawyers Need to Know

Generative AI refers to artificial intelligence that can generate new content such as text and images. And as you can imagine, this has far-reaching implications for everything in the legal industry, from the future of work to specific practice areas. And, as is often the case with any new AI technology gaining traction, many headlines questioned whether lawyers were becoming obsolete. No, but this new technology poses many interesting questions and opportunities for our industry. Lindsay O’Connor, Head of Core Products for LexisNexis Pacific, recently spoke with Greg Dickason, Managing Director of LexisNexis, about how he sees these changes.

“From my point of view, the hype around generative AI comes from what I call the ‘gradual change’ of technology.” Greg says.

“Fundamentally, the technology has now reached a stage where it can do much more useful and impressive things than it ever has. We quickly saw the potential of the search engine to rank and display information, and as a result, people never looked back.”

Netflix and smartphones are also step-change technologies that have a fundamental understanding (via AI) of how users want to interact with each other and have evolved human behavior almost instantly. Generative AI is no different. Tools like ChatGPT fundamentally rewrite the rules of human-AI interaction (just kidding). Instead of searching for a term and having the relevant results returned by an algorithm, that algorithm now has a say. You can ask questions and refine your answers in exactly the same conversational way you would when talking to others.

“ChatGPT has gone through hype cycles at a very fast rate,” says Greg. “But most people have tried it before and know it has limitations. I think it does, because they know what it can and can’t do.”

Greg points out that the real difference between the new model underpinning the generative AI platform cohort and the previous model is the amount of training data available. He points out that there are very apt parallels between what we see now and our own human development.

“These are neural networks, and humans are neural networks. What parents know is that children grow in stages. And to some extent, that’s what we’re seeing here.”

LexisNexis has been working and experimenting with various AI models in its products since 2018. This includes Google’s models and his earlier GPT models (such as GPT3), leveraging the best of these technologies for legal solutions. “Fundamentally, law is centered around language,” Greg says. “That’s why we can quickly determine that this technology will impact our legal practice.”

With a large LLM in law, the benefits are immediate. The first is reviewing, drafting, and editing by competent AI. LLM also has the ability to change the tone and even match the tone to an existing document. Rapidly change the way attorneys conduct research by returning the most relevant results and highlighting the relevant parts within them, as well as providing an overview or summary of why a particular result is relevant I can. Even the basics of client communication, such as sending letters and emails, can be greatly streamlined with the help of generative AI. AI requires human oversight, but the benefits are not in eliminating jobs, but in process efficiencies, which are very valuable.

“You’d be hard-pressed to find someone in the legal industry who wouldn’t want to spend more time in their day,” says Greg.

But there are limits to what this technology can do. The first and most commonly discussed accuracy score is around 75-80%, but this is certainly not enough for lawyers to place too much trust in the output without thorough consideration. The second is currency, says Greg. “In other words, given that the latest GPT model is complete at the end of 2021, there is no data for 18 months, so if you ask a question about current affairs, the answer is no. It’s very important,” he said, adding that the current headline models are trained on public datasets, meaning they don’t have access to vast amounts of subject-specific data behind paywalls and firewalls. also mentions

Looking to the future, Greg is both optimistic and cautious about the potential integration of this exciting technology into more legal solutions.

“I think there will be great technology coming out soon, but in the short term, I think a lot of startups will take advantage of it to provide easy point solutions that will ultimately disappoint. In the future, some companies will share their developed solutions with their customers.In our case, we used both technologies. and We leverage the content strengths we have built over decades to create solutions that are much more robust and actually meet customer needs. “

Greg says the real key for companies looking to harness the power of AI most effectively is putting their content in a format and location that AI can easily consume. The advantage of using a purpose-built solution like LexisNexis is that data hygiene is already done. All that remains is for the user to become familiar with the tool and start working. Greg says this will be a key edge in the future.

“These capabilities are emerging rapidly, but that doesn’t mean they’re going to destroy law firms, but that those firms that utilize these capabilities more effectively will become more competitive. And there lies the challenge.”

This article is an excerpt from Episode 1 of the AI-Decoded Legal Talks podcast series.Listen to the full episode here

At LexisNexis, promoting the rule of law is our purpose, and this purpose guides our actions, so when developing solutions, all development is guided by RELX’s responsible AI principles. Make sure it’s aligned. relx-responsible-ai-principles-0622.pdf. These principles ensure that our solutions are developed in line with our values ​​and maintain our position as a thought leader in the market.