ChatGPT, Cheating, and Change: What All Educators Need to Know, and What's Next

• In Event: Ethics of AI in Ed: Five Humans and ChatGPT Discuss

Abstract

Author: Presenter is an assistant professor in the Computer Science and Engineering department at a research one, AAU University in the South East of the United States.

Purposes:

The presenter aims to critically examine the current state and future implications of Natural Language Processing (NLP), particularly focusing on large language models (LLMs) such as GPT-4 and ChatGPT. The goal is to understand the technology's impact on the education sector, notably the emerging problems of cheating, and propose strategies to navigate these issues.

Theoretical Framework:

The presentation approaches the subject from the dual perspectives of computer science and education, integrating understandings of AI capabilities and the nature of learning and assessment in educational settings.

Methods, Techniques, or Modes of Inquiry:

The presenter employs an analytical and comparative method to evaluate various claims surrounding LLM capabilities, including its potential misuse for cheating in academic contexts. This analysis is juxtaposed with an explanation of NLP evolution, specifically autoregressive models and pre-training plus fine-tuning techniques.

Data Sources, Evidence, Objects, or Materials:

The presentation draws on a variety of data sources, including technical literature on NLP and LLMs, case examples of AI use in education, and observations about the rapid evolution of AI capabilities and their misuse.

Substantiated Conclusions:

The presenter argues that the undetectable misuse of AI tools for academic cheating poses a significant challenge to educators. This problem is amplified by the ease of access and rapid improvements in AI technology, outpacing the capabilities of detection tech. However, the presenter also recognizes the potential benefits of embracing such technology in education, suggesting that educators modify their assessment techniques and course designs to account for AI's presence.

Scientific Significance:

This presentation offers a unique and timely analysis of the intersection of advanced AI technology and educational ethics. It brings to light the pressing issue of academic integrity in the AI era, contributing to the broader discussion of AI's role in education. The insights and recommendations provide valuable starting points for educators, researchers, and policy-makers to address and manage the challenges presented by AI tools in the academic sphere.

Author

• Gene Louis Kim, University of South Florida