Student Perspectives on Using a Large Language Model (LLM) for an Assignment on Professional Ethics

Overview

• The paper discusses how the rise of Large Language Models (LLMs) has sparked discussions among educators on how they will impact curricula, assessments, and student competencies.
• It also explores the ethical questions and concerns that generative AI and LLMs have raised for computing educators and professionals.
• The paper presents an assignment within a course on professional competencies, including ethics, for computing master's students.
• The assignment involved students discussing an ethical case study, using an LLM as a discussion partner, and reflecting on the experience.

Plain English Explanation

As large language models (LLMs) have become more advanced, educators have started to think about how they might affect the way students learn and are evaluated. These powerful AI systems have also raised ethical questions for people working in the computing field.

In this study, the researchers gave students in a master's-level course an assignment related to professional ethics. The students were asked to discuss a fictional scenario where a researcher is deciding whether to attend a conference in Hawaii. They used an LLM as a partner to help them explore the ethical considerations involved.

The students found the LLM to be a useful tool for expanding their knowledge and understanding of the case. However, they also recognized that the LLM suggested some unrealistic options. Ultimately, the students had to choose the course of action themselves, which the researchers saw as a positive outcome.

The researchers noted that some students didn't fully perceive the LLM as a tool, but rather as an authoritative source of information. This suggests that educators may need to provide more guidance to students on how to effectively use LLMs, especially when it comes to practicing critical thinking around ethical dilemmas.

Technical Explanation

The paper presents an assignment within a course on professional competencies, including ethics, for computing master's students. The assignment involved three main tasks:

1. Participating in in-class discussions on a case study about a fictional researcher considering whether to attend the real CHI 2024 conference in Hawaii.
2. Using an LLM of their choice as a discussion partner for the case study.
3. Documenting both the in-class discussions and the discussions with the LLM, and reflecting on their experience using the LLM.

The students reported positive experiences with the LLM, noting that it helped increase their knowledge and understanding of the case. However, they also identified limitations, as the LLM suggested some unfeasible options. Importantly, the LLM did not select a course of action, so the students had to choose for themselves, which the researchers saw as a coherent outcome.

The researchers observed that some students did not perceive the LLM as a tool, but rather as an authoritative source of knowledge. This suggests a need for more instruction for students on using LLMs, particularly when it comes to practicing critical thinking around ethical concerns related to LLMs and navigating their moral implications.

Critical Analysis

The researchers acknowledge that their study is limited in scope, as it only involved a single assignment within a single course. They also note that the students' experiences may have been influenced by their prior knowledge and familiarity with LLMs.

One potential concern is that students may become overly reliant on LLMs as a source of information, rather than developing their own critical thinking skills. The researchers emphasize the need for careful guidance and instruction to help students use LLMs effectively as tools, rather than treating them as authoritative sources.

Additionally, the paper does not delve deeply into the specific ethical considerations or dilemmas that were explored in the case study. It would be interesting to see a more detailed analysis of the ethical issues raised and how the LLM and students grappled with them.

Overall, the study provides a useful starting point for understanding how LLMs can be integrated into computing ethics education. However, more research is needed to fully explore the long-term implications and best practices for using LLMs in this context.

Conclusion

This paper presents an interesting approach to integrating large language models (LLMs) into a computing ethics course. The researchers found that students generally had positive experiences using an LLM as a discussion partner, but they also identified the need for more guidance and instruction on effectively using these tools.

The findings suggest that LLMs can be a valuable resource for expanding students' knowledge and understanding of ethical dilemmas, but educators must be careful to ensure that students develop their own critical thinking skills and do not simply rely on the LLM as an authoritative source.

As the use of LLMs in education continues to evolve, this study offers important insights for computing educators and highlights the need for further research on the ethical implications and best practices for integrating these powerful AI systems into the classroom.