The concept of digitalisation and Artificial Intelligence (AI) are often confused in the everyday language. When we talk about digitalisation, we usually mean the replacement of traditional case management systems supported by paper-based processes and isolated Information Technology (IT) systems with various electronic means, sub-processes and systems that are connected in real time. Online forms instead of paper documents and applications. Automated, online communication with chatbots comes to the front instead of human interface.
Artificial Intelligence is already moving to the next level and is transforming the IT as a service unit into a collaborative partner. With the help of AI and a wide range of databases, legal cases can be modelled and analysed in a very short period of time. The results of this research can then be used in legal decisions.
II. The Signs
I think there is no doubt that something is happening in the IT industry but we’re just scratching the surface. Everyone talks about the new achievements, the progressive and very tempting future. However, there are voices warning humanity to take a step back and recalibrate the goal and the scope of different Legal Technologies, like for example Machine Learning.
I am sure that most people reading this article have already heard about the legal case in Colombia where the judge involved ChatGPT in the decision-making process in a case of an autistic boy. Since many blogs have been written about the revolutionary ChatGPT and its impact on providing legal advice, this time in my article I would like to focus on the possible use and impact of AI on judicial work.
III. The Columbian Case and ChatGPT
In the Colombian case – as described in The Guardian article – the judge used the AI tool, in particular ChatGPT to decide whether an autistic child’s insurance should cover all of the costs of his medical treatment. The judge argued that Colombian law is often very complex and sometimes it is really hard to navigate but ChatGPT could help to make the right decision. During the process the judge also used precedents from previous rulings to support his decision. The judge also wanted to make it absolutely clear that that the decision was made by himself not by the ChatGPT. He declared that he used the ChatGPT because of its advantages which are fast and comprehensive information processing and data mining. He also stressed out that he believes ChatGPT and other similar IT systems can assist judges in their work but will never replace them.
During his decision making process and before making his final decision, the judge asked the chatbot whether or not an autistic minor could be exempted from paying for his therapies. The ChatGPT responded by agreeing with the judge’s statement and confirmings that minors diagnosed with an autism are exempt from paying fees for their therapies. This case has raised a lot of discussion about the use of AI in law and has been criticised by many legal experts. But before we declare that AI is from the devil let us take a closer look at the case.
IV. Benefits, Advantages and Disadvantages
Before we go any further we need to make it clear that it was the judge who made the decision not the AI. The AI only helped the judge to get an overview of relevant cases and legal regulation. In this case the AI was used as a source of big data information that has helped the judge to be in line with all the current legal regulations and precedents. So, this means that one of the possible uses of AI in the judicial work is
data analysis and data processing.
Artificial intelligence can be used to process large amounts of past judgments in just a few minutes which would otherwise take months or years. This leads us to the next benefit of the of AI in the judiciary, which is
better time management, shorten time scales, faster judgements.
I think it is very important to see that in this case the helping hand came from a chatbot (ChatGPT) and not from a more sophisticated data mining IT solution that would have been based on thousands of judgements and court cases. The Colombian chatbot story was a perfect example for us to see how widespread the use of AI is and how AI could help us in our work. I am sure that AI could be a very effective tool in the work of the judiciary. I do not mean in general that AI should replace human judges or that AI should make decisions instead of judges. But I believe that in certain cases (e.g. misdemeanours) AI could handle the case more effectively and could be a real alternative to humans. In these cases, the main role of the judges would be to appeal against the AI’s decision. This solution would bring the simple and smaller cases to a much
quicker conclusion and reduce the workloads of the courts.
In the case of more complex matters, the AI function would have provided a rich source of information on similar cases and decisions to help the judges make their judgements. In my opinion, this leads to completely new requirements for judges. Knowledge of previous cases and judgments would be less important in the selection of judges, as the AI could reach a conclusion within seconds from studying thousands of cases. Therefore the main factor of a sophisticated judge would be the practical approach to making judgments using AI and incorporating the human factor into the verdict.
This direction leads us to the next milestone of AI in the judicial work.
If the AI is used to provide an authentic and complex source of past judgments and legal cases, then judgements will be
more reasoned and informed
based on thousands of previous judgments and legal cases. If all judges are working from the same databases and using the same technology to support their work, then the advantage is that the results of these judges will be parallel because they are using the same methods, and there is unlikely to be any difference between the decision of similar cases whether they were made in different parts of a country. This means that the
judge’s personal judgement rather than lack of knowledge would determine the final decision.
Besides the decision making, the AI could also have an impact on the work of courts in other ways, however although this option can also be a bit dangerous. Just as in the case of data mining of thousands of legal cases there are thousands of judgements. This means that with the use of AI it would be
very easy to profile judges and to predict what kind of judgement can be expected from a judge.
From this point it is only a small step to a marketplace of judges where we can find all the information on judgements, legal cases and judges. This possibility could make it possible to decide which
judge should hear our case if we want a particular judgement.
I think that this profiling should not be made public because it could lead to harmful behaviour towards judges. People could have access to the decisions of the judges and they could come to conclusions that are not fair or accurate. That’s why I’m of the opinion that this profiling should be within the confines of the court. This would be a good tool within the court to assess the performance, the professionalism and the quality of the work of the judges, helping to determine which judge needs to be motivated and which one should be rewarded.
V. Near Future
I think that in the near future AI will be used in more and more areas of judicial work. But in the case of tools that can help us to work more efficiently, it is a perfectly obvious thing. The use of modern technology avoids the possibility of dysfunctions of human mind and emotions and therefore put our lives on a more predictable track. The legal profession is no different.
In my opinion, we should leave it to the AI to make the decision more often in the case of simple legal cases. In these cases (e.g. misdemeanours) the AI itself, rather than a judge, could make the decision. Therefore, the use of AI would help to focus judges on monitoring and reviewing the suggestions made by the AI and to focus on more specific and complicated cases.
There are so many several precedents in the judicial work which are mainly forgotten or are not taken into account. The use of AI as a data mining tool could help to solve this problem and to provide a broader view of the same cases and legal decisions helping to make informed decisions.
I think that AI will never completely replace the human being in judicial work. This is simply because there is the human factor that allows the judge to decide what sentence should be applied in a particular case. The AI would not be able to make such a complex decision. So, I am sure that the AI will is going to crawl into judicial work but we have to be careful how far we let it in.
VI. Near Future
In this article, I have tried to show the potential use of AI in the work of the judiciary. Although this is a very complex area, we can already see some parts of this profession where the use of AI could make a difference and may bring many benefits. Overall, I think AI will never replace judges. It can help them to be more efficient, more effective and more accurate. But we should not forget that we have to keep an eye on the AI, because it is serving us, not us serving it.