GPT-4 AI Chatbot Achieves Higher Scores Than Prospective Lawyers on Legal Ethics Exam
In a groundbreaking study, OpenAI’s GPT-4 chatbot has surpassed the performance of aspiring lawyers on a legal ethics exam. According to a report released by LegalOn Technologies, the artificial intelligence (AI) software answered 74% of the questions correctly on a simulated Multistate Professional Responsibility Exam (MPRE), surpassing the estimated national average of 68% for human test takers. This achievement highlights the potential of AI in assisting lawyers with ethical compliance and aligning with their professional responsibilities.
The study conducted by LegalOn Technologies, a provider of AI software for contract reviews, suggests that in the future, AI could play a crucial role in helping lawyers navigate complex ethical issues. By achieving exceptional scores on the MPRE, GPT-4 showcases its ability to interpret and apply black-letter ethical rules as effectively as aspiring lawyers.
GPT-4, a generative AI technology developed by startup OpenAI in partnership with Microsoft, has garnered attention for its human-level performance on various professional and academic benchmarks. The software, released in March, is a large multimodal model capable of processing both text and images to generate accurate responses.
While GPT-4 showed remarkable performance on the legal ethics exam, it is important to note that the National Conference of Bar Examiners, responsible for developing the MPRE, has expressed reservations regarding the study’s claims. A spokesperson for the organization stated that attorneys possess a unique set of skills that AI currently cannot match.
The MPRE is a crucial requirement for prospective lawyers in almost every state, except Wisconsin, as it evaluates their understanding of topics such as conflicts of interest, lawyer-client relationships, and confidentiality. GPT-4 showcased its proficiency in various areas, excelling in questions related to conflict of interest (91% correct responses) and lawyer-client relationships (88% correct responses). However, its accuracy dipped slightly in questions concerning communications about legal services (71% correct responses) and safekeeping funds (72% correct responses).
The impact of AI on legal education and attorney licensure has been the subject of ongoing research. Previous studies have indicated that earlier versions of GPT-4 achieved passing scores on law school final exams and even simulated bar exams. However, while GPT-4’s access improved the speed of law students’ legal writing assignments, it did not enhance the quality of their work.
As the legal profession continues to embrace technological advancements, the integration of AI poses new opportunities and challenges. While GPT-4’s exceptional performance in legal ethics is a significant milestone, it is vital to recognize that the unique skills and expertise of human attorneys are irreplaceable. The coexistence of AI and legal professionals can potentially result in more efficient and compliant legal practices.
In conclusion, GPT-4’s ability to outperform prospective lawyers on the legal ethics exam brings to light the promising potential of AI in the legal field. Although AI can provide valuable assistance with ethical compliance, lawyers’ skills remain incomparable and essential. The evolving nature of the legal profession will undoubtedly continue to incorporate technology, transforming legal practices and benefiting both legal professionals and clients alike.