In recent years, we have seen a wide use of Artificial Intelligence (AI) applications in the Internet and everywhere. Natural Language Processing and Machine Learning are important sub-fields of AI that have made Chatbots and Conversational AI applications possible. Those algorithms are built based on historical data in order to create language models, however historical data could be intrinsically discriminatory. This article investigates whether a Conversational AI could identify offensive language and it will show how large language models often produce quite a bit of unethical behavior because of bias in the historical data. Our low-level proof-of-concept will present the challenges to detect offensive language in social media and it will discuss some steps to propitiate strong results in the detection of offensive language and unethical behavior using a Conversational AI.

This article describes a methodology named Causal Analysis and Resolution (CAR) based on Goals, Questions, and Metrics (GQM) principles. Indicators are defined based on metrics for a decision-making process. Its main contributions are the construction of an information process system model and a prototype, involving GQM approach, in a quantitative definition and qualitative metrics. The CAR methodology is a process area (PA) of the Capability Maturity Model Integrated (CMMi) for software development from Carnegie Mellon University. This PA was used to eliminate systematic error cases listed in a Technical Report (TR) generated by CAR. An information system model was created to allow the elimination of defects, errors, and failures in a design pattern named IO Manager, during the test phase, and before its publication in a components library. The prototype was created using Rational Rose RealTime (RRRT) with focus on verification tests. It provided a quality assessment to the IO Manager design pattern. The use of this methodology was based on GQM and CAR along with the information process system model. The developed prototype aimed to monitor errors on design pattern tests in real-time embedded system of a software production line.