Professor Fails Entire Class Due To False Claims Made By ChatGPT

Artificial intelligence (AI) has rapidly advanced and found its way into various aspects of our lives, including education; while AI has the potential to enhance learning experiences and streamline processes, its ethical implications and proper use remain critical considerations.

Recently, an incident at Texas A&M University–Commerce brought these concerns to the forefront when a professor mistakenly accused an entire class of plagiarism based on the use of an AI language model called ChatGPT. This incident has sparked discussions about the responsible use of AI in education and the need for clear guidelines to prevent such misunderstandings.

The professor (Dr. Jared Mumm) utilized ChatGPT to test whether his students had employed AI to generate their final assignments, however, he was unaware that ChatGPT does not function as a plagiarism detection tool, and after running the papers through ChatGPT, the AI mistakenly claimed that all the assignments were produced by the chatbot itself.

Consequently, the class faced allegations of plagiarism, leading to the temporary withholding of their diplomas. Dr. Mumm failed every essay with an “X” grade, giving the students the option to submit a makeup assignment or risk failing the course and not graduating. Some students attempted to prove the authenticity of their work by providing timestamps on their Google Documents, but the professor dismissively responded, “I don’t grade AI bullshit.”

Although one student managed to clear their name by providing Google Docs timestamps and receiving an apology, the matter has been escalated to the university’s administration. Texas A&M University confirmed that no students failed the class or were prevented from graduating due to the incident. Investigations are currently underway, and the students’ diplomas are being withheld pending further action.

The misuse of ChatGPT in this case highlights the need for proper understanding and awareness of the capabilities and limitations of AI tools in educational settings. While ChatGPT can generate text, including collegiate-level essays, it is not designed to detect AI-generated plagiarism.

Tools for detecting plagiarism

AI programs specifically developed for detecting plagiarism include Winston AI, Content at Scale, Writer AI, GPTZero, and Giant Language Model Test Room (GLTR). OpenAI, the parent company of ChatGPT, does offer its own plagiarism detection tool — but its accuracy is considered limited.

This incident has sparked debates surrounding the responsible use of AI in education and calls for clearer guidelines on the use of AI tools to avoid similar misunderstandings and unfair accusations. While some schools in the United States have already taken steps to block ChatGPT from being used on campuses, the response from institutions globally remains varied.

It is essential for educational institutions to carefully consider the implications of AI usage and establish appropriate policies to ensure academic integrity while leveraging the benefits of AI technology.

Paulo Montenegro
Categories: Robots
Tags: AI (Artificial Intelligence), ChatGPT