Emily D. Carpenter

    This practicum project adds a data engineering contribution to the GENESIS (General Emergent Norms, Ethics, and Societies in Silico) research project at Regis University. The GENESIS project is investigating whether intelligent computer agents are able to develop cooperative behaviors and moral reasoning via simulated interactions, specifically through use of the Iterated Prisoner's Dilemma (IPD) testing environment.

    The core simulation framework and Large Language Model (LLM) agent code has been provided by the project researchers, but there is no systematic method to store, organize, and analyze experimental results from multiple game runs. As such, this project addresses that gap by designing and implementing a database solution to capture experimental data generated by intelligent agents playing the IPD game in addition to studying the possible emergence of cooperation between agents.

    The infrastructure created in this practicum project will enable researchers to conduct analysis across experiments to identify the conditions that lead to cooperative emergence, examine patterns in agent reasoning, and contribute towards planned research publications, the first step in a multi-year plan for the GENESIS research project. This practicum demonstrates practical data engineering skills including database design and interface with the research project, version control, and documentation practices while contributing to ongoing Artificial Intelligence (AI) ethics research that explores the boundaries between optimization and moral behavior.

    This project continued data engineering contributions to the GENESIS (General Emergent Norms, Ethics, and Societies in Silico) research project at Regis University. The GENESIS project investigates whether intelligent computer agents are able to develop cooperative behaviors and moral reasoning through simulated interactions, specifically through the use of the Iterated Prisoner's Dilemma (IPD) testing environment.

    The core simulation framework and the large language model (LLM) agent code have been provided by the principal investigators of the research project, and a persistent data storage solution was developed in the first practicum course (MSDS 692). This phase of practicum study containerized the research platform for portable deployment to allow the principal investigators to collaborate with academic partners in the field of Data Science to contribute to the ongoing Artificial Intelligence (AI) ethics research that explores the boundaries between optimization and moral behavior.