Matthew Golab is the Director of Legal Informatics and R+D at Gilbert + Tobin. He leads a specialized in-house multidisciplinary legal informatics team that utilizes a variety of data analytics and eDiscovery, and other AI technology tools. Matthew has more than 20 years of experience in the legal technology industry, including two of Australia’s preeminent law firms.
If you’ve been in the legal profession for any amount of time, then I’m sure at some point, someone will have told you that there are nuance problems, it’s very different, and things are difficult to change. Whether you buy into that across the entire legal tech ecosystem, looking at problem-solving in legal when it comes to AI, there are some unique challenges. And I’m excited about this episode because we explore some of those challenges today. Before we get started, do note that during the conversation with Matthew, we jumped straight into the deep end, right at the beginning of the conversation.
However, both Matthew and I make a concentrated effort to de-mystify as much of the jargon as we go through. And during the episode, we really do tackle why are artificial intelligence problems so difficult to solve when it comes to Legal use cases? Why don’t we have better-performing models? Why don’t we have models that actually work on any number of different data types or different types of documents or languages? Especially if you look at things like what’s happening in the consumer world, or if you look at even other business verticals where things just seem to be much much further ahead. Why does it feel that the legal profession is still a long way behind?
We tackled that and a lot more. I don’t promise that we have all of the answers, but certainly, we have a healthy discussion into why things are the way they are and what needs to happen to bring about change in the future.
Before we dive in, if you haven’t subscribed to my newsletter, please do so. As part of the newsletter, I actually provided some additional commentary to supplement this episode, specifically discussing how things like synthetic AI are being used in other industries where there are similar challenges in the form of small data sets with little variations.
- Introduction to the podcast (0:28)
- The deliberate use of language in legal vs. general language training sets (4:51)
- NLP, NLG, and NLU (7:08)
- Overview of the AI development process (9:49)
- The deliberate balance to be achieved when introducing technology to lawyers (13:22)
- High-risk tolerance as a barrier to continuous learning (15:09)
- The general sentiment about AI and the future (17:08)
- The challenge of working with different jurisdictions and languages (22:34)
- Training a specialist system vs a general-purpose system (26:37)
- The biggest improvement in tech within law firms (29:17)