TLDR: Soppia is a structured prompting framework designed to bring consistency and transparency to the assessment of non-pecuniary damages in personal injury cases. Developed by Jorge Alberto Araujo, it uses AI to guide legal professionals through a multi-criteria analysis, employing a calibrated scoring system with dual logic and a weighting mechanism. The framework classifies damage severity and suggests compensation ranges, aiming to reduce subjective variability in judicial decisions and promote explainable, auditable legal AI. It is adaptable across various legal contexts and jurisdictions.
The legal landscape often presents a significant challenge in consistently applying complex rules, particularly when it comes to determining compensation for non-pecuniary damages. These damages, which include pain, suffering, and emotional distress, lack a clear monetary equivalent, often leading to subjective judicial discretion and considerable variability in outcomes for similar cases. This inconsistency, referred to as “noise” in human judgment, can undermine public trust and predictability in legal decisions.
Introducing Soppia: A Framework for Fairer Assessments
To address this critical issue, Jorge Alberto Araujo introduces Soppia—a structured prompting framework designed to assist legal professionals. Soppia, which stands for System for Ordered Proportional and Pondered Intelligent Assessment, leverages advanced AI to ensure a comprehensive and balanced analysis of all stipulated criteria. The goal is to faithfully fulfill the legislator’s intent that compensation be determined through a holistic assessment of each case’s specific circumstances.
The framework is not intended to replace human judges but rather to augment their decision-making process with a transparent, consistent, and auditable tool. Its name, ‘Soppia,’ reflects its core values of simplicity, approachability, and reliability in supporting complex judicial decisions.
How Soppia Works: A Structured Approach
Soppia’s methodology is built upon three fundamental pillars: decision-making theory, explainable artificial intelligence (XAI), and prompt engineering. It aims to reduce the variability in judgments by structuring the decision-making process, making it a form of “decision hygiene.”
The framework operates through four main components:
-
Identification and Definition of Criteria: The first step involves identifying legally relevant criteria, which can be derived from statutes, case law, or legal doctrine. As a case study, Soppia utilizes the 12 criteria for non-pecuniary damages outlined in Article 223-G of the Brazilian Consolidation of Labor Laws (CLT).
-
Calibrated Scoring System with Dual Logic: Each criterion is evaluated on a scale of 1 to 5 points. A key innovation is the dual-logic system: for most criteria, a higher score indicates greater severity (direct logic), while for mitigating factors, a lower presence of the factor results in a higher score (inverse logic). This ensures that a higher score consistently reflects greater damage severity.
-
Weighting System: Recognizing that not all criteria hold equal importance, Soppia incorporates a weighting system. For instance, the possibility of recovery and the duration of effects receive higher weights due to their decisive role in distinguishing temporary from permanent injuries. The final score for a case is calculated as the weighted sum of the scores for each criterion.
-
Classification and Final Adjustment: The total weighted score then classifies the damage into one of four severity categories: Mild, Medium, Severe, or Very Severe. Each category corresponds to a predefined compensation range, often expressed as a multiple of a relevant baseline, such as the victim’s salary. A fine-tuning mechanism allows for further modulation within the selected range for a more granular recommendation.
The entire methodology can be encapsulated in a detailed prompt for a Large Language Model (LLM), guiding the AI to perform a step-by-step analysis and generate a transparent, well-documented output. For a deeper dive into the framework, including the complete prompt and implementation materials, you can refer to the research paper here: Soppia: A Structured Prompting Framework.
Broad Applications and Reproducibility
While the initial case study focuses on Brazilian law, the Soppia framework is highly adaptable. Judges can use the generated reports to structure and justify their decisions, while lawyers can assess case strength and formulate settlement strategies. Companies can employ it as a risk management tool, and academics can use it for legal analysis. The framework can be readily modified for other legal contexts, such as consumer law or environmental law, by adjusting the criteria and weights to reflect local legal systems and precedents.
Also Read:
- HIKMA: Advancing Scholarly Communication with AI-Powered Conferences
- EGO-Prompt: Automating LLM Adaptation for Specialized Tasks with Evolving Domain Knowledge
Conclusion: Towards Explainable and Equitable Justice
The Soppia Framework offers a practical and methodological contribution to assessing non-pecuniary damages, paving the way for more consistent, transparent, and well-founded judicial decisions. By combining structured legal criteria with AI’s analytical power, it demonstrates that judicial “noise” can be significantly reduced without sacrificing human oversight, explainability, or the necessary individualization of analysis. Soppia aims to ensure that technological advancements in law always serve the fundamental principles of justice: fairness, proportionality, and rational justification.
