The following are the outputs of the captioning taken during an IGF intervention. Although it is largely accurate, in some cases it may be incomplete or inaccurate due to inaudible passages or transcription errors. It is posted as an aid, but should not be treated as an authoritative record.
***
>> RAFAL WIECZERZAK: I would like to start today's panel where we will talk about AI in the Courts as an Opportunity for Economic Proceedings. Today, we would like to discuss the potential this technology in the judiciary. Mainly from the perspective of the benefits it can bring to business.
On the opportunities and challenges of using artificial intelligence in business proceedings, I will be discussing today with Anna Pietruszka, expert in business models based on enable intelligence and artificial intelligence.
Gabriela Bar, doctor of law, attorney at law, specialist in the area of new technology law with particular focus on artificial intelligence.
And Robert Sebinski attorney at law, specialist in litigation, including commercial disputes, arbitrations and mediations proceedings.
A judge in the civil department, district Court Konrad Wasik. In Poland, member of the working group on artificial intelligence Ministry of Digitalization.
My name is Rafal Wieczerzak. It is my privilege to moderate this panel. Today with me the panel is moderated online with Magdalena Golonka. I would like to warmly welcome our guests who have decide to join us. I believe our discussion will be interesting. And allow us to answer the questions, the questions that is the essence of the panel.
Before we start, I would like to inform you today, our panel is only 30 minutes, so be concise.
The first question I would like to ask you what business expect from the judiciary in the 21st Century.
>> Anna Pietruszka: To answer the question, we have to understand where the business is right now. Business globally, in my opinion, in the transition between the fourth and fifth Industrial Revolution. The fourth brought us digitalization. The focus of the fifth Industrial Revolution is a human‑centric approach as well as it should be AI driven. This is where the business sits. Let's see where the Courts globally jurisdiction is. In many places, it is only the very early beginnings of the fourth revolution. So I think the main expectation would be to fulfill the technology gap between the Courts and the business.
Why it is important? It is important because the business it is nothing new to what business expects from the Court. Business expects justice. Nothing new here. But what the justice means nowadays for the business? It is the value delivered on time, follow the law and human rights and all of the regulations. But time is crucial here.
So currently, you know, the dialogue between the business and courts is difficult because of the technological gap. And challenges it brings to the communicate. It really brings the frustration I guess on both sides. But certainly on the business side.
>> RAFAL WIECZERZAK: Thank you for your input.
Very interesting. I would like to confront your point of view with Konrad. Konrad, how do you think if the judges also feel frustrated by the fact that in their work, besides the need to focus on the ruling on serious cases a number of additional activities must be performed?
>> Konrad Wasik: Good morning everyone. It is a pleasure to connect with you all remotely for this panel. European justice systems face a number of problems. The judiciary is not keeping up with modern technologies. It struggles with rapidly changing law and judges lack quick and efficient action. We may say resolving a case is like crossing the finish line, but before that, there is an entire race and the necessity to deal with many formalized and time‑consuming activities. I would like to point that it is not only about the number of cases the judge have, but also about many other tasks required by civil procedure. We may say the more activities needed and the higher is the probability it takes a court to reach a decision. To put that in context, what I mean. I mean we work on paper case files. We don't have electronic files. We have to write justification of decisions in majority of cases. In civil procedures, it is a request and the fee is small. And manual searching for paper documents in huge volumes of case files. Manual searching for judgments in other cases. There is also public expectancy that the judge have to decide about everything from court fees to formal documents.
There is circulation of traditional correspondence, we have to analyze if everyone involved in the case is properly notified. We have time‑consuming hearings and many, many more. So the problem is that resolving a case is a marathon, but the business expects a quick race.
According to the latest report made by European commissions that compares judiciary system across the European Union, it is estimated time needed in Europe to resolve contested civil and commercial case, I draw your attention to the word "contested" we have dispute and many parties, documents, arguments. At first instance, in 2021, ranges from 100 to more than 700 days per case. In Poland it is about three hundred days per case. In my opinion, it is way too long for modern business Sector. And AI may change it, but it has to be done responsibly. Thank you.
>> RAFAL WIECZERZAK: Thank you, Konrad. It seems like the judges are interested in implementing this technology. Robert, how do you see the possibility of implementing AI based technology into the Courts from the legal perspective?
>> Robert: Thank you for your question. Artificial intelligence AI is already present. It is currently being used in various aspects of legal system, including in courts. However, AI doesn't replace human judges. But supports them in order to improve efficiency and help decision‑making. The extent and role of AI implemented in the Court process can significantly depending on jurisdiction. Legal traditions of course ethical considerations.
AI is usually being utilized in courts in the following area. At the pre‑court stage for predictive analytics. At the Court stage for case management and support. An example of a court powered by AI is the Internet court in China, operating since 2017. In this court, artificial intelligence eliminated time‑consuming communication process and is using blockchain technology as one of the digital technologies to secure of electronic evidence. One of the tools is a robot. I apologize for the pronunciation (?) which answers questions about technical matters. The robot can help individuals to navigate the Court platforms. The Internet court primarily established to handle Internet related legal cases and disputes over e‑commerce. In 2018, due to rapid growth of e‑commerce and online activities, two more online courts were established in China. In Beijing and Guangzhou. In Guangzhou, average time the Internet court for considering a case is 38 days. In Beijing, 40 days.
The second example of an Internet court that I would like to mention is the Civil Resolution Tribunal, CRT operating in Canada since 2016. This quasi court is a part of the judiciary of province of British Colombia, Canada. A very interesting digital tool provided by CRT is solution explorer, self‑help tool. The explorer asks simple questions to provide personalized legal information and analyzes based on the data provided by users, using the explorer is free and anonymous. According to CRT, the satisfaction survey from August this year, 92% of respondents found CRT staff to be professional. 87 found the CRT online services not difficult to use. 92% respondents agreed that CRT provided information that prepared them to resolve disputes. And 82% felt that this tribunal treated them fairly throughout the process.
Finally, maybe a fly in the ointment. In the case of China's online courts, they're independent and impartiality is questioned. Because the I.T. solutions and tools are often prepared by technology companies or subsidies arc affiliated with parties to the disputes.
The same technology companies collect and process huge amounts of data of participants in online court processes.
>> RAFAL WIECZERZAK: You mention the technical issue in the handling of the private data of entities. I would like to stop here and ask Gabriela about ethical aspects. How do people see AI and how to build trust with the technology?
>> Gabriela Bar: Hi everyone. It is a difficult question for 2:00 in the morning. This is the time in Poland. I will try, I will do my best. People's perception of AI I think varies on their exposure and understanding with the technology. To build the trust in AI, especially in sensitive areas like the judiciary, it is crucial to ensure transparency, fairness and reliability in the AIs is.
Studies conducted this year in Poland indicate that the lack of trust toward AI system increases as the decision made by AI is more directly to individuals.
So it is easy for people to accept AI forecasting the weather. But there is resistance whether AI decisions related to access to public services, for example.
Of course, the other issue is we humans, we're biased very often. And this bias can be transferred to AI systems. So if an AI system is trained on bias data it can perpetuate and amplify the biases. And it leads of course to discriminatory decisions.
This illustrates the importance of transparency in AI systems. And it contains also audits, creating ethical guidelines. Maintaining open channels for communication to address concerns which are essential. Another dimension to this is the belief that in some areas, especially judicial proceedings, we require so‑called life experience and empathy. That kind of quantity that AI currently lacks.
And I think personally, that it isn't an unfounded belief, at least for now. Because human judges don't only apply the law, they often interpret it, based on the unique circumstances of each case, based on their life experience, empathy and while AI systems may process information rapidly and recognize patterns in data, we probably still need this human touch so probably even if we integrate AI into other judicial system, we still should use human factors, let's say. And for now, I think it is a condition for the trust in AI systems. Thank you.
>> RAFAL WIECZERZAK: That you, Gabriela. Business has a lot of experience with creating and using AI. Mostly of generating profits, finding new clients, growing the business. And I would like to come back to you Anna Pietruszka, what lessons has business learned for implementing AI based innovations. Where can we start when implementing this in courts?
>> Anna Pietruszka: Few things from business to mention here. First of all, whatever technological change we implement, we shouldn't lose the value proposition. Every single business has a business model and in the center of the model is the value proposition. So if we do want to transform the cards and want to implement technology, we have to do it in the way that the value proposition delivered by the Courts is not lost.
The value proposition in its essence is justice, simple as that.
I think first of all, it has to be secured by the Governments and the different institutions for instance, for example, in Europe, as you may know, right now, there is this big AI Act, and I'm keeping my fingers crossed for Spain, who is actually leading this project. As you know, their presidency will be finished in European Union and will be finished soon. So I'm really hoping they will make sure that this one is done.
And for I think in AI Act, it should be secured that the value proposition is there. This is lesson number one. The second lesson, again, nothing complicated. We as human being, we don't like changes. This technological revolution has to be done step‑by‑step.
The great lesson from business is the hybrid intelligence driven business models.
I strongly believe that in courts it should be done based on the hybrid intelligence model. The hybrid intelligence is the blend of artificial intelligence and human intelligence. It has to be done in the way that this blend is implemented step‑by‑step to be successful.
I think those are the two biggest lessons that we can bring from the business to the Courts.
>> RAFAL WIECZERZAK: Thank you, I agree with you that we should start implementing AI with the artificial intelligence model. There is such a kind of AI as explainable artificial intelligence. Gabriela, do you think that just such artificial intelligence should be used in court?
>> Gabriela Bar: Yes, of course. I would say that focusing on the explainable AI would be especially important in the judiciary. Decisions if the judicial system have profound authority, not just for businesses but for society in idea. It is imperative the AI tools used in this sector should be transparent. But the decision‑making process should be understood and reviewed. So XAI provides a way to make the workings of AI more accountable which aligns with the principle of justice and fairness. Again, coming back to the ethical principles that are so important when using this technology.
The purpose and the scope of legal or ethical and legal requirements, for AI in this context may differ of course, depending on the profile of recipients and the level of the available explanation. Because probably we need different explanation for the expert. For the judge, for lawyers. And different for end users affected by the decisions made by AI. And for that kind of end user. The most important will be justification of the decision to let the people, this person to appeal for example.
So the concept of explainable AI, it is not solely about understanding the data used. The way it is applied in neural networks or facts which the AI system considered when assessing a particular situation, a particular case. But at its core, it is about the human need for justice and societal acceptance. So when a court delivers a verdict, it is not just a cold calculated decision based on data and law. I hope so at least. There is a decision that must be socially acceptable and perceived in line with the social values. The human factor in judicial decisions is crucial.
So I think that the understanding which is behind every case the story which is behind each case, the people which are involved, it should be also the part of explainable AI in judiciary. Thank you.
>> RAFAL WIECZERZAK: Thank you. You mention how the artificial intelligence system should look like in the judiciary. However, commercial disputes arise not only in the courtroom. But also in the form of arbitration of mediation. Robert is there a place for artificial intelligence in alternative dispute resolution methods?
>> Robert: Definitely, Rafal Wieczerzak. In my opinion, the era of automation is coming to an end in the alternate dispute resolution area. Artificial intelligence should play a dominant role in ADR techniques.
In the case of nonbiding resource, occurred out of this Resolution such as mediation or negotiations. Replacing the mediator or Moderator with an algorithm in my opinion is not associated way lack of trust in AI. To the same extent as when replacing a human judge with machine.
Personally, I often have more reservations about traditional mediations because I notice that the mediators goal is to achieve the settlement itself. Not the first settlement. Therefore, if the computer Programme is additionally equipped with digital tools predicting possible solutions, to the dispute, I believe, that it may inspire greater trust among the parties of the conflict than the human. Personally, I'm a supporter of developing a conciliation based on artificial intelligence.
I mean, presenting nonbinding solution through a computer Programme or application instead of human conciliator, the proper algorithm would present a nonbinding solution proposal after analyzing the facts and evidence presented by the parties. And assist the parties in reaching a voluntary settlement. In my opinion, most parties of the conflict want to find out whether they are right at the very beginning. Before they decide to engage in costly court disputers.
‑‑ disputes. Especially in commercial matters.
The basic question the attorney hears from his client is ... please, tell us what our chances of winning are.
My take‑home message is that the conciliation Programme powered by AI may be an excellent tool for silencing the economic disputes.
I believe that this type of Programme could help to avoid clogging up the judiciary. Thank you.
>> RAFAL WIECZERZAK: Thank you very much Robert. I would like to address my last question to Konrad. Konrad, what do judges expect from AI? And what are the concerns of the judicial community and courts regarding the implementation of this technology?
>> Konrad Wasik: Thank you for this question. I believe that AI is a great chance to improve courts. The judicial community has both concerns and expectations. There is a concern about bias issues. AI might enhance biases if it is trained on bias data. The data is crucial. There is also a problem that AI might not fully grasp, fully catch nuanced human judgments, emotions, ethical considerations, which are important in sentencing, which Gabriela noticed. There are also concerns about transparency of AI predictions, about data privacy. About data security.
Yeah, I also ask my Secretary what is the concern of ‑‑ from the Court staff perspective. And there is also a concern about job losses due to automation. There is also a risk that judges might rely too heavily on AI, which may lead to lack of critical thinking.
I think there are more concerns, so in my opinion it is important to understand that AI should not replace a human judgment. From my point of view, we need to use AI to deal with simple, time‑consuming, and administrative tasks. Only then will we see good effects we can move boundaries and to use AI to dist in sending, in small‑claims procedures and maybe in online court. We need to know that if you want to promote trustworthy AI, there is no place for errors.
Because if people loses trust in AI, it will be very difficult to justify its use in the judiciary. That is why it is so important to introduce AI initially in small steps. I would like also to emphasize that this is a perspective of our AI work group. We had prepared a report with recommendations on the use of AI in the Polish judiciary. The report includes a number of interesting implementations on how to improve the polish justice system. It will be available on the home page. But when we talk about expectations, I think that AI should be used to automate routine tasks. For example, it can analyze electronic case files, analyze receipts if everyone is noticed. Speech to text recognition, which replaces the necessity to write the protocol. The anonymization of rulings that office staff have to do it manually. Translation during hearings, translation of documents because we have to involve language expert every time if you want to translate document or hear a witness that doesn't speak Polish. Analyze huge amounts of data. AI can analyze huge amount of data and present a judge with list of similar cases and verdicts. It can be used for case management in the Court. We can implement chat bots to provide legal information to the parties. To sum it up, I see AI as a future of the Polish justice system. There is a place between collaboration between the man and machine. I think it can bring benefits to both the judiciary system and society as a whole. Europe has the chance to promote trustworthy AI with new regulations. In Poland we are also moving in the direction to use modern technologies for people. I think we have to do it in small steps in responsible way.
Finally, I would like also to emphasize that according to the report made by the European bank for construction and development and the Global law firm, Poland is ready for the introduction of online courts. Thank you.
>> RAFAL WIECZERZAK: Thank you very much. Unfortunately, we must end our panel. But please, feel free to ask question to our speakers as we have limited time. Our speakers will answer questions after the panel. Questions can also be asked via chat on the Zoom platform. Online participants are encouraged to send their questions to the dedicated email address IGF@CK‑legal.PL. And now we have a question. So ...
>> ATTENDEE: Hi. I'm Gopal Krishna. Senior advocate, the President of Nepal Bar Association. My question is a reason is the main tool to provide justice. Isn't AI contradict with the justice making? In my opinion, in the substantial issue, in my opinion, it can help the justice making process. But it cannot make help to the substantial issue. This question is for the panelists. Thank you very much.
>> Anna Pietruszka: I will answer in a few sentences and we take it offline because of the time limitation. I think when we talk about transition ‑‑ technology transition and the changes for technology, we have to really think about single roles. So how does single role will change. How are daily activities, how the judge's activities will change. That is really important to understand. And only from this little activities we can answer activity by activity whether or not we can deliver the justice with the technology usage. And to design it in a way that the justice can be delivered as the value proposition. That is the most important. That is the lesson from the business as well. To do it really step‑by‑step by funk, by single activity. How the single activity will change in the process. And only then we can do it in the way that the value will be delivered.
We are running out of time. We have the information here. I'm happy to take more questions offline. Thank you so much for your question.
>> RAFAL WIECZERZAK: Yes, we must end. I will summarize our panel. During the discussion we cover a wide range of topic considering the implementation of AI in judiciary including technical legal aspects. We explore areas in which AI could prove to be an effective addition to the judicial system. I generally hope today panelists encourage the reflection of adoption of technology within the judiciary. Thank you very much. The last information we will respond to all email questions within 14 days. Thank you very much.
(Thank‑yous)