Evaluation of AI Ethics Tools in Language Models: A Developers' Perspective Case Study

Abstract

In Artificial Intelligence (AI), language models have gained significant importance due to the widespread adoption of systems capable of simulating realistic conversations with humans through text generation. Because of their impact on society, developing and deploying these language models must be done responsibly, with attention to their negative impacts and possible harms. In this scenario, the number of AI Ethics Tools (AIETs) publications has recently increased. These AIETs are designed to help developers, companies, governments, and other stakeholders establish trust, transparency, and responsibility with their technologies by bringing accepted values to guide AI's design, development, and use stages. However, many AIETs lack good documentation, examples of use, and proof of their effectiveness in practice. This paper presents a methodology for evaluating AIETs in language models. Our approach involved an extensive literature survey on 213 AIETs, and after applying inclusion and exclusion criteria, we selected four AIETs: Model Cards, ALTAI, FactSheets, and Harms Modeling. For evaluation, we applied AIETs to language models developed for the Portuguese language, conducting 35 hours of interviews with their developers. The evaluation considered the developers' perspective on the AIETs' use and quality in helping to identify ethical considerations about their model. The results suggest that the applied AIETs serve as a guide for formulating general ethical considerations about language models. However, we note that they do not address unique aspects of these models, such as idiomatic expressions. Additionally, these AIETs did not help to identify potential negative impacts of models for the Portuguese language.

7 figures, 11 tables. Accepted for publication in AI and Ethics

References in corpus (30)

• Artificial Intelligence: the global landscape of ethics guidelines
• Model Cards for Model Reporting
• Explaining Machine Learning Classifiers through Diverse Counterfactual Explanations
• Worldwide AI Ethics: a review of 200 guidelines and recommendations for AI governance
• Explainability Fact Sheets: A Framework for Systematic Assessment of Explainable Approaches
• The AI Assessment Scale (AIAS): A Framework for Ethical Integration of Generative AI in Educational Assessment
• Auditing large language models: a three-layered approach
• BOLD: Dataset and Metrics for Measuring Biases in Open-Ended Language Generation
• Foolbox: A Python toolbox to benchmark the robustness of machine learning models
• Interactive Model Cards: A Human-Centered Approach to Model Documentation
• Charting the Sociotechnical Gap in Explainable AI: A Framework to Address the Gap in XAI
• Sabiá: Portuguese Large Language Models
• Zeno: An Interactive Framework for Behavioral Evaluation of Machine Learning
• LiFT: A Scalable Framework for Measuring Fairness in ML Applications
• Fairness Score and Process Standardization: Framework for Fairness Certification in Artificial Intelligence Systems
• Amazon SageMaker Clarify: Machine Learning Bias Detection and Explainability in the Cloud
• Advancing Neural Encoding of Portuguese with Transformer Albertina PT-*
• The Dataset Nutrition Label (2nd Gen): Leveraging Context to Mitigate Harms in Artificial Intelligence
• Augmented Datasheets for Speech Datasets and Ethical Decision-Making
• XAutoML: A Visual Analytics Tool for Understanding and Validating Automated Machine Learning
• Crossing the principle-practice gap in AI ethics with ethical problem-solving
• DC-Check: A Data-Centric AI checklist to guide the development of reliable machine learning systems
• POLARIS: A framework to guide the development of Trustworthy AI systems
• DIANES: A DEI Audit Toolkit for News Sources
• Model-Agnostic Private Learning via Stability
• A toolkit of dilemmas: Beyond debiasing and fairness formulas for responsible AI/ML
• Responsible and Inclusive Technology Framework: A Formative Framework to Promote Societal Considerations in Information Technology Contexts
• The BIG Argument for AI Safety Cases
• Towards Responsible AI Music: an Investigation of Trustworthy Features for Creative Systems
• Sabiá-3 Technical Report