First-year law students will be all too familiar with the complexities of deciphering a court judgment. Even lawyers with decades of experience will acknowledge the high degree of interpretation that is needed to understand a judge’s legal opinions and extract key information from them. Surely, no algorithm could perform this complex human-centered task and even use its analysis to predict case outcomes? Jozef Maruščák, the Founder of CourtQuant, has shown that with access to the appropriate data and with the right model, it most certainly can! 

As a law undergraduate at the University of Cambridge, concerned by the challenges faced by litigants to accurately estimate their chance of winning their case, the costs,  and duration of the process, Jozef saw an opportunity to leverage the power of AI to make the justice system and the litigation process more transparent. But creating software with the ability to predict the outcome of a case is easier said than done. In his conversation with Annabel, Jozef breaks down the major hurdles they faced: 

Data – Accessing and structuring it

In order to build an algorithm, you need access to a robust data source, which in this case would mainly consist of court judgments. Accessing these in the UK is very difficult and in some European countries it’s impossible. Assuming you can get your hands on the data, you then need to structure it. This means working the data so that key information is extracted from inconsistent judgments in a standardized and automatic fashion.  

Training models

A model that finds relationships between various data points (e.g. what was the case about, who were the lawyers, the parties, what were the costs etc…) and the outcome, also need to be developed. Based on these identified patterns, the system will learn and can then be used to predict future case outcomes. 

Persuading the industry to trust the tool

For the very traditional and conservative legal industry, it is counter-intuitive to believe that data science could ever solve a legal case. Although there is a real need for the problem CourtQuant addressed to be solved, the reality that technology is the only way to do so has not unilaterally sunk in yet. 

“Right now, there are only a handful of lawyers in the whole of the UK who actually believe that you can rely on an AI system to tell you the result of a trial”.

With the automation of certain legal processes and the increasing adoption of LegalTech improving and speeding up a lawyer’s job, a shift in mindset will come. But it takes time. 

“The future of this world lies in technology”.

Interested in hearing more about Jozef’s experience launching his entrepreneurial venture whilst still at uni? Want to learn more about his current role as Head of Business Development at the software engineering consultancy Sudolabs? Give the podcast below a listen now!

The Wired Wig explains helps to plug in Digital and Technology Law into businesses. It explains technology concepts and informs business leaders and students how the Law could respond to innovation.

The podcast is available on Spotify, Google Podcasts, Apple Podcasts and other podcast providers found on Anchor.