The Roman goddess Justitia – here in front of a courthouse in Hamburg – is the symbol of justice and symbolizes the incorruptibility and independence of the court. Nevertheless, the emphasis on all factors still plays a role in case law – AI is not in a position to do this. Even if all the technical and human requirements were met, the formidable robot judge would currently have no chance of taking up his duties before the German courts. “There are several provisions in the Basic Law that exclude a decision by a machine,” explains Isabelle Biallaß. “In addition, there are several ordinary legal provisions – for example in the German Judges Act – from which it follows that the judge must be a human being.” A judge ruling in Germany must also have German citizenship – a criterion that an AI cannot meet. But this is not the only obstacle to the expansion of forensic AI: at present, the necessary technical and human structures are still lacking. However, the legal framework would also need to be adapted, which would require changes to the rules of procedure and procedure – always taking into account applicable national and constitutional law. The biggest challenge, however, will likely be to bring about a change in the digital mindset and ensure that the public rejects their suspicions of artificial intelligence in the courtroom. In the perception of many people, technological progress and digitalization are a double-edged sword. On the one hand, machines need to relieve as many unloved tasks as possible, make everyday life easier, and speed up processes. On the other hand, when it comes to machines taking control of living and working spaces, there is concern about an alleged loss of control associated – this is shown, for example, by the suspicion of autonomous vehicles. To learn more about Legal Tech Fellows, visit the DLA Piper Careers blog.
An article on this topic has already been published in the blog “Technology`s Legal Edge”. Tools can be easily integrated into existing workflows. “The use of AI in the judicial system is still in its infancy,” explains Isabelle Biallaß. The judge of the Essen District Court sits on the board of directors of the EDV-Gerichtstag e. V., who has been involved in the digitalisation of the judicial system for 30 years and was responsible for the topics of the use of AI in the judicial system and legal technology as a consultant for the Ministry of the Land of North Rhine-Westphalia. “In the field of judicial decision-making, AI systems are not yet used on a daily basis,” she says. However, the judiciary is currently examining in several pilot projects how the use of AI in the legal system could be significantly expanded. Thanks to our many years of experience in the field of legal document automation, we also advise our clients if they wish to automate the types of contracts frequently used in theirs in order to relieve their legal department of the corresponding requests. Among others, the Organisation for Economic Co-operation and Development (OECD), which advises Nicole Formica-Schiller as an AI expert and established rules for artificial intelligence in 2019, addresses this largely hypothetical question. A year earlier, the European Commission for the Efficiency of Justice (CEPEJ) had published five principles of the Ethical Charter on the use of artificial intelligence in and its environment. For the use of AI in the legal system, they require respect for fundamental rights, user control, transparency – as well as impartiality and fairness – quality and safety, and non-discrimination. The effort required by manual intervention decreases with each iteration.
Artificial intelligence (AI) is on the rise. The number of applications of Legal Tech AI is also constantly increasing in the legal sector. As one of the pioneers of legal technology solutions in the legal market, CMS uses machine learning document analysis software that helps us analyze contracts and other legal documents efficiently and quickly. In order to be able to automatically create a wide variety of contract documents, we have been relying on contract automation for years. Digital input masks and meticulously prepared template automation make it easier for our lawyers to create a first draft that maps the constellation correctly and consistently in every detail and in which all boring work such as grammatical adjustments are implemented with a single click. Rapid dissemination of new clauses in the event of changes to the framework conditions. In a proposal to regulate AI, launched by the European Commission in April 2021, the use of powerful AI systems in the judicial system is classified as high-risk. This means that it would have to be subject to particularly strict control for this to happen. This is justified by the potentially significant impact on democracy, the rule of law and individual freedoms. The so-called AI law also mentions the possible threat to an impartial tribunal – but this only applies to certain AIs. Purely related administrative activities that are currently being expanded in Germany are not classified as high risk. Does the European Commission therefore expect artificial intelligence to be vested with decision-making power before European courts in the distant or near future? For example, the research project “Legal Analysis” of the Bavarian State Ministry is currently working on the possibility of an anonymous database of court decisions that could make it possible to analyse the decisions of future legal cases – especially in civil law.
AI is also used in law enforcement: since 2020, the federal government`s KISTRA research project has been developing AI methods for automated detection of crimes such as hate crimes on the Internet. Elsewhere, AI combs through large amounts of data in search of child pornography. Systems that read legally relevant information from electronic documents or read documents submitted to the court are also being tested. The workload associated with repetitive but necessary document checks (e.g. NDA) is decreasing. The full article is available on beck-online (paid). Before a trial can begin, many administrative tasks must first be completed and often large amounts of data must be processed – processes that are time-consuming and partly responsible for the backlog of cases in some areas of justice. “In this respect, the use of AI can certainly lead to relief for employees and a `purification` of procedures,” says Nicole Formica-Schiller, Member of the Board of Directors of the KI Bundesverband and CEO and founder of Pamanicor Health AG. In the German legal system, AI performs the tasks of assistants and assists lawyers by collecting, processing and preparing data. The expansion of these tasks is slow because machine learning applications need to be fed with a lot of data before they can be used. In the case of legal technology, AI learns from judgments.
In fact, there is no shortage of them – but very few are public. The reason for this is data protection. In particular, reviewing legal documents is extremely time-consuming and expensive. Good lawyers are in demand, but supply is scarce. The use of artificial intelligence as a medium is therefore obvious. Processes are accelerated by support and intuitive operation. The idea that the algorithms and computer programs of our legal systems can make decisions that directly affect our lives and freedoms sounds like something out of a science fiction noir movie. But artificial intelligence in the judicial system is not a dream for the future: legal technology exists and has long found its way into courts and prosecutors` offices – also in Germany. “Iudex non calculat?” ALA calculate! Our AI experts have developed a tool to support their own legal department in the preliminary review of confidentiality agreements. The automated legal assistant (ALA) is an internal tool that allows you to capture and classify texts, highlight critical points and propose changes via a uniform interface. Three of our legal experts manage more than 1,600 projects and 2,100 contracts per year. The aim is not to relieve lawyers of decisions, but rather to relieve them of repetitive, shortened and simplified processes through digital support and useful functions.
Focus: The review of texts in legal documents such as confidentiality agreements, NDAs or general conditions.
