DOI: 10.31038/ALE.2024123

Abstract

Using generative AI (Chat GPT3.5), the paper shows how a legal case such as eviction for non-payment of rent can be ‘brought to life’. The first simulation shows how AI can be used as a preparatory device for the Mind Genomics experiment, thus making the lawyer’s job easier when using Mind Genomics for a legal case. This first simulation focuses on AI-created questions and answers, and the ‘guesstimate’ by AI as to the likelihood that a jury will find ‘for the plaintiff’ vs ‘for the defendant’. The second simulation focuses on different reasons for the eviction, and AI-based analyses of legal aspects vs ethical aspects. This second stimulation demonstrates the ability of AI to consider the nuances of a case. In both simulations the focus is on types of issues that are best described by phrase ‘subjective, and open to interpretation.’

Introduction

As legal education evolves, law schools are placing greater emphasis on equipping students with the skills to critically analyze intricate legal cases. Traditional methods for teaching law methods frequently rely on rote memorization of statutes and procedures. However, a growing number of educators are now embracing experiential learning opportunities, allowing students to hone their analytical skills in practical contexts. In this new era of legal education, innovative technologies such as generative AI are increasingly being utilized to simulate legal cases, enabling students to navigate real-world scenarios effectively. When combined with the emerging science of Mind Genomics the new mix enables students, or indeed anyone, to gain a human perspective on the law, viz., put a human face on the facts through simulation. The simulation allows learning by asking ‘what-if’ questions, not about the case law itself, but about the softer, human aspects [1-5].

This paper shows two simulations of the same case, where a landlord seeks to evict tenants who have fallen behind on rent for two months due to job loss. The case itself is rich with legal versus emotional and moral complexities. In the classroom, students might end up relishing the back-and-forth analysis, as they engage in a spirited discussion, exploring the pertinent legal principles, the rights of both landlords and tenants, and possible solutions to the issue at hand. This exchange of ideas enables students to examine various viewpoints, sharpen their arguments, and gain a richer insight into the intricacies of legal decision-making.Can we make this back and forth into an AI application, using currently available LLM’s, large language models? As this paper shows, we can do so in ways which are staggeringly fast, inexpensive, scalable, and adaptable to many situations [6-8].

Simulation in the legal field breathes life into cases, offering a vivid experience which textbooks and lectures often fail to convey. Generative AI lets law students explore the intricate complexities and subtleties of real-world legal dilemmas through simulated cases. In this scenario, the AI offers clarity on landlord-tenant laws, the eviction process, and the delicate balance of interests between property owners and tenants.Students can delve into various viewpoints and possible results of the case through simulation. The legal rights and responsibilities of both landlords and tenants can be examined, alongside an evaluation of the ethical implications stemming from the decisions made by each party. This practical method sharpens critical thinking and fosters a richer comprehension of legal principles at work.

This paper explores the two simulations, based on the same ‘facts of the case’. Simulation 1 deals with what to present to a jury to get the jury to vote for the client. Simulation 2 moves more deeply into the role of the lawyer as an advocate for their client.

Simulation 1: Mr. Owner vs Mr. & Mrs. Tenant – Rent Dispute and the Search for Discovering Powerful Messages Resonating with the Jury

We begin with the facts in the case, shown in Table 1. The stimulation is set up to provide messages that can be tested by the emerging science of Mind Genomics, to identify those specific messages which will resonate with a jury [9-12].

Table 1: Facts of the case and instructions to AI (Simulation 1)

The actual Mind Genomics experiment with real people (not shown) follows a straightforward process. The steps are as follows:

1. Select a topic
2. Develop four different questions about that topic which give a human face to the topic
3. For each question develop four different answers to the question, phrased as stand-alone phrases which paint a word picture
4. Combine the 16 elements into small, easy-to-read vignettes comprising a minimum of two and a maximum of four elements. The composition of the 24 vignettes to be evaluated by a single respondent follows an underlying experimental design. Each question contributes only one element to the vignette. Across the set of 24 vignettes each of the 16 elements appears 5x and is absent 19x. Mathematically the structure of each respondent’s set of 24 vignettes is identical, but the permutation ensures that the vignettes are different across all respondents.
5. Each respondent evaluates a permuted set of 24 vignettes, using a simple rating scale. The vignette is a stand-alone composition, easy to read. The respondent requires about 3-4 seconds to scan the combination of messages and assign a rating.
6. The data are put into a simple data base to prepare them for statistical analysis. Ordinary least-squares, dummy variable regression analysis shows the driving power of each element, first for each respondent, and then for groups of respondents defined by methods such as cluster analysis.

Up to now the creation of raw material, questions especially, has emerged as a stumbling block. For whatever reason, people are intimidated. Over a 24-year period, 1998 to 2022, the same plaint was heard, always by the researcher using the tool. It was simply daunting, although children had fewer problems than did adults, and especially far fewer problems than did professionals. Professionals wanted perfection in the vignette, ending up enamored with overly polished, wordy paragraphs, hard to read, hard to test more than a few.

The introduction of AI in 2022 revolutionized this emerging science of Mind Genomics. As Table 1 showed it is straightforward to come up with the statement of the ‘facts of the case’. It is the creation of the questions and elements (answer) which are difficult.

Table 2 shows what emerged after one iteration requiring 10-15 seconds. These four answers are statements about the case, and the AI’s ‘guesstimate’ about the jury’s decision based on each answer. The key benefit to critical thinking is the ability of AI to jump-start the thought process. To ‘have fun’ one could iterate, with each iteration requiring the 15 seconds. An encyclopedia of questions and elements (answers) would emerge in about 10 minutes or shorter. The enterprising user could change the facts in the case as well and explore different ‘what if’ aspects, using the Idea Coach features.

Table 2: Simulation of four questions and four elements (answers to each question), prepared for a Mind Genomics experiment with actual people (Simulation 1).

The effort is finished, whether one iteration or 100 iterations, the results of each iteration are stored on an Excel spreadsheet, with one tab dedicated to each iteration. Thus, it is possible to run many iterations, acquire the data, but leave the subsequent analysis to an off-line effort, that effort taking about 5-6 hours. The results are returned to the user by email. The remainder of this simulation is generated after the study preparation is closed. AI then goes over its own questions and answers and provides a summarization and deeper analysis shown in Table 3, and beyond. Table 3 presents the first set of information,

1. Key Ideas: AI can provide valuable insights into key ideas by analyzing large amounts of data and identifying patterns and trends. It can summarize complex concepts and present them in a clear and concise manner. This can help individuals better understand key concepts and make informed decisions.
2. Themes: AI can also identify common themes across different sources of information, allowing for a deeper understanding of the subject matter. This can help in identifying trends and patterns that may not be apparent at first glance.
3. Perspectives: AI can provide different perspectives on a topic by analyzing a wide range of sources and presenting diverse viewpoints. This can help individuals gain a more comprehensive understanding of the subject matter and consider different angles.
4. What is missing: AI can also highlight what is missing in a particular discussion or research study by pointing out gaps in information or potential areas for further exploration. This can guide researchers in identifying areas that need further investigation.
5. Alternative viewpoints: Additionally, AI can present alternative viewpoints on a topic by analyzing conflicting information and presenting different sides of the argument. This can help individuals consider different perspectives and make well-informed decisions.

Table 3: AI summarization of simulation 1 in terms of key ideas, themes, perspectives, what is missing and alternative viewpoints.

Table 4 compares interested audiences versus opposing audiences regarding the material presented in Table 3. AI can expand the topic by questions and answers tailored to specific interested audiences, allowing for a more personalized and engaging experience for learners. AI can create content that is relevant and meaningful content. increasing their motivation and interest in the material. Going a step further, AI can also identify prospective opposing audiences by presenting diverse perspectives and challenging viewpoints. By generating questions and answers that provoke critical thinking and debate, AI can stimulate discussions and encourage individuals to consider alternative viewpoints. This can lead to a more well-rounded understanding of the topic and foster a culture of open-mindedness and tolerance.

Table 4: AI synthesis of interested versus opposing audiences, based on material in Table 3.

The final analysis as of this writing (November 2024) is the suggestion of innovations. The initial suggestion of innovations produced a list of AI-generated innovations. During the summer of 2024, the Mind Genomics system was modified so that the post-creation of the study (not-yet-run with people) would offer a far more complete analysis of each innovation that was being created and offered to the user. These nine aspects appear below and are shown in Table 5 for each of the four innovations.

1. Specific suggestion: AI can provide specific and targeted suggestions based on data analysis and patterns
2. Explanation why the suggestion is relevant:
3. Importance and uniqueness of the suggestion:
4. Social Good resulting from the suggestion.
5. Slogan which emblemizes the suggestion:
6. Investment pitch:
7. Potential investor pushback:
8. Answer to the potential investor pushback:
9. Compromise, go forward positioning:

Table 5: The four innovations suggested by AI and an AI ‘work-up’ of each innovation

Simulation 2 – Dealing with the Same Issue, but From the Point of View as the Lawyers Helping Both Sides as ‘Trusted Parties’

In this simulation we move backwards, away from the situation of a lawyer trying to anticipate what to say to the jury (Mind Genomics approach), to understanding the ‘facts in the case’ from the point of view of the client. Here the lawyer becomes a trusted friend. The process becomes more personal and empathic.

Table 6 shows the ‘facts of the case’ and the instructions to AI to generate the relevant questions and answers. Table 7 shows nine different questions and the structured answers to those questions.

Table 6: Facts of the case and instructions to AI (Simulation 2)

It is worth noting here that generative AI such as Chat GPT3.5 used here, can be instructed to simulate different people talking to each other, e.g., listening in as a ‘fly on the wall’ to the lawyer’s conference.

Table 7: Nine relevant questions generated by AI regarding the case, with answers simulated by AI with respect to from the legal standpoint and the moral standpoint.

Discussion and Conclusion

Simulations are crucial for students to understand the intricacies of legal decision-making, enabling active engagement with complex scenarios and practicing their knowledge in real-world contexts. These simulations enhance students’ understanding of the law and its real-world uses, as well as their critical thinking and problem-solving abilities. The future of law schools will likely employ diverse techniques and technologies, guiding students to think critically about complex issues. Mind Genomics, enhanced by AI, presents a groundbreaking solution for the legal system, offering students an engaging and interactive approach to grasp the intricacies of legal cases. This technology has the potential to reshape the legal landscape, fostering quicker case resolutions, enhancing transparency, and expanding access to justice for everyone. Mind Genomics and generative AI can craft realistic legal scenarios by examining the many factors which shape legal decision-making, including case law, evidence, and ethical considerations. These scenarios can provide students with practical experience to enhance their legal reasoning skills in a simulated environment.

Mind Genomics, powered by AI, has the potential to revolutionize the legal profession by improving decision-making, streamlining case resolutions, and ensuring justice is served efficiently and transparently. This technology would benefit students by providing interactive learning tools, real-world case studies, and hands-on experience, while professors would have access to cutting-edge research tools and real-time data. The implementation of Mind Genomics could transform the legal practice, leading to more efficient case resolutions, improved transparency, and greater access to justice for all individuals.

Reflection and debriefing sessions on simulated legal cases can help students improve their legal reasoning abilities, bridge the gap between theoretical legal knowledge and practical application in real-world scenarios, and develop a deeper understanding of human factors involved in legal decision-making. At the same, it might well be productive is during the application of AI to legal issues the user requests both legal and ethical/moral considerations as separable steps in the process.In conclusion, using technology-driven simulations in legal education can promote active learning, enhance critical thinking skills, and provide practical experience in legal decision-making, equipping students for successful careers in law.

Acknowledgments

Author Moskowitz wishes to acknowledge the help of the audience at his lecture at the law school in of the University of Kragujevac, November, 2024. Many of the points of view regarding ethics and morality and the law were challenged by the students and professors, leading to refinements in the authors’ thinking.

The authors would like to thank Vanessa M. Arcenas and Isabelle Porat for help in preparing this manuscript for publication.

References

1. Curcio AA (2009) Assessing differently and using empirical studies to see if it makes a difference: Can law schools do it better. Quinnipiac L Rev 27: 899.
2. Horwitz P (2012) What Ails the Law Schools. Mich L Rev 111: 955.
3. Katz HN (2006) Evaluating the skills curriculum: Challenges and opportunities for law schools. Mercer L Rev 59: 909.
4. Mertz E (2007) The language of law school: learning to” think like a lawyer”. Oxford University Press, USA.
5. Shah M (2010) The legal education bubble: how law schools should respond to changes in the legal market. Geo J Legal Ethics, 23: 843.
6. Alarie B, Niblett A, Yoon AH (2018) How artificial intelligence will affect the practice of law. University of Toronto Law Journal, 68: 106-124.
7. Jinzhe Tan, Hannes Westermann, Karim Benyekhlef (2023)ChatGPT as an Artificial Lawyer? Canad Workshop on Artificial Intelligence for Access to Justice (AI4AJ 2023)
8. Susskind R, Susskind RE (2023) Tomorrow’s lawyers: An introduction to your future. Oxford University Press.
9. Moskowitz H, Kover A, Papajorgji P (2022) Applying mind genomics to social sciences. IGI Global.
10. Moskowitz HR, Wren J,Papajorgji P (2020) Mind genomics and the law. Mauritius: LAP Lambert Academic Publishing.
11. Papajorgji P (2023) Knowledge as a service: The case of Mind Genomics. EuroMediterranean, 19: 34-47.
12. Porretta S, Gere A, Radványi D,Moskowitz H (2019) Mind Genomics (Conjoint Analysis): The new concept research in the analysis of consumer behaviour and choice. Trends in food science & technology, 84: 29-33.