We have seen significant progress in using artificial intelligence for legal research applications to promote efficiency and reduce costs.
While AI in its current form augments the skill set of lawyers in streamlining the research process, human supervision and verification is still needed to ensure relevance and accuracy in legal research output. When discussing the use of AI in legal research, it is important to distinguish between generative AI and extractive AI.
Generative and extractive AI – for lawyers the difference between the two is significant
Generative AI, being the concept behind well-known platforms such as ChatGPT, cannot guarantee that it won’t “hallucinate” and on occasions produce fictitious answers as fact. One story involving two New York lawyers illustrates the risks of current generative AI in legal research. A New York Court sanctioned the lawyers after they submitted case law returned by a generative AI platform to the Court, which was later found to have been entirely fabricated by the AI platform.
Extractive AI, on the other hand, refers to an algorithm that extracts data points from the data it has been trained on, rather than generating new outputs.
For example, LexisNexis’ new generative AI tool, Argument Analyser, harnesses extractive AI technology to analyse a passage of text from which it identifies relevant legal issues. On this basis, it makes authoritative suggestions as to relevant case law and legislation. There is little risk in harnessing extractive AI for legal research, provided that the data on which the AI is trained is authoritative.
How AI beats traditional research methods
We recently held a tech-a-thon with LexisNexis to launch Argument Analyser. The AI tool can navigate a vast case law database, retrieve pertinent case recommendations, analyse cited cases and legislation, and create a map of key legal issues based on a segment from a legal document.
The session pits lawyers into two teams – one using traditional legal research methods and the other deploying Argument Analyser. Participants conducted typical research tasks, which included finding and downloading legislation and case law cited in the hypothetical submissions and locating relevant case law for reply.
Teams quickly realised the benefits of AI in legal research, with the Argument Analyser side achieving accurate results much quicker than the team engaging with ‘traditional’ legal research. Overall, the session outlined the current capabilities and future potential of AI in legal research.
Future of legal tech
AI is already integrated into most of the major legal research platforms used by law firms in Australia.
We are witnessing a growing appetite for new technologies that save time, reduce cost, and increase efficiencies. Some recent developments include:
In July this year, LexisNexis launched Lexis+, a new, next-generation, AI powered legal information platform based on interconnected experiences that saves time and surfaces key insights to help mitigate the risk of missing something important. LexisNexis is now working on launching Lexis+ AI, a tool that includes generative AI solutions for conversational search, insightful summarisation and intelligent drafting.
Thomson Reuters recently announced their acquisition of Casetext, Inc., an artificial intelligence and machine learning company focusing on AI technology solutions for the legal industry. Casetext’s crown jewel is the platform CoCounsel. This AI-powered legal assistant seeks to help lawyers save time and money by automating contract and document review and fast-tracking legal research through generative AI.
Jasmine, powered by BarNet Jade, is a conversational search engine leveraging AI technology to reduce the time lawyers spend finding authoritative case law. While its development is ongoing, the platform harnesses cutting-edge natural language search techniques that helps it to understand the research question at a deeper level and, in turn, Jasmine can return results that are more relevant than a basic keyword search.
Greg Dickason, MD LexisNexis Asia Pacific is a firm believer that “AI is a powerful tool to supplement ‘traditional’ legal research and confirm its completeness. With prudent supervision and verification alongside ‘traditional’ forms of legal research, this technology can significantly reduce the time spent by lawyers and law firms in gathering what they need driving efficiency and effectiveness across the entire legal ecosystem”.
Authors: Keren Smith, Chief Knowledge Officer, and Lachlan Ahale, Undergraduate, at Holding Redlich
The information in this article is of a general nature and is not intended to address the circumstances of any particular individual or entity. Although we endeavour to provide accurate and timely information, we do not guarantee that the information in this article is accurate at the date it is received or that it will continue to be accurate in the future.