Thursday, December 26, 2024
The Weekly: Get top MTSU stories in your inbox by subscribing to The Weekly, a Sidelines newsletter delivered each Wednesday.

MTSU professors cannot always detect AI — what now?

Date:

Share post:

Featured Photo by Rolf van Root/Unsplash

Story by Matthew Giffin

Middle Tennessee State University professors will not be able to report students for academic integrity violations based only on generative AI detector reports, according to an email to faculty from the school’s director for academic integrity obtained by MTSU Sidelines.

“Since the reliability of the AI detector is in question, my office will not be accepting cases based solely on the AI detector report,” Director of Academic Integrity Michael Baily said in the email. “I recommend that faculty members speak to students and perhaps have students rewrite assignments. Of course, grades are always determined by the faculty member.”

OpenAi, the creators of ChatGPT, removed its generative AI detection tool from the internet due to a “low rate of accuracy” in July, according to their website. Baily also cited concerns about AI detectors being biased against non-native English speakers.

OpenAI is still working on tools to detect AI-generated content, according to their website. Some computer science experts, however, are skeptical.

“I don’t think it will ever be possible to make a foolproof AI writing detector,” said MTSU professor John Wallin, the director of the school’s computational and data science Ph.D. program, in an email to Sidelines.

“We need to be holding the line”

MTSU currently has no university-wide statement or policy about the use of generative AI for class assignments.

“There’s kind of a move to restrict it, to restrict usage of it, and that’s not gonna work,” MTSU associate professor of interactive media Todd O’Neill told Sidelines about faculty members’ attitudes toward generative AI. “And we all need to really embrace it.”

Professors’ policies about generative AI use vary from class to class at MTSU. Wallin created a list of recommendations and sample policies for faculty to use for their courses. In the list, he included policies for faculty who wish to prohibit all uses of generative AI or want to allow it under varying levels of restriction.

“We need to be holding the line on how this kind of technology gets used,” O’Neill said about generative AI use. “And when I say, ‘We,’ I mean students and faculty.”

Because no guaranteed technological solution exists to detect AI-generated content, some faculty are rethinking how they will assess their students in the future to circumvent the possibility of students using tools like ChatGPT to write their answers for them.

In-class work and discussion may need to weigh heavier on grading than in the past, O’Neill said. “Even, like, D2L discussions and things like that, they’re not great measures.”

MTSU’s “AI Champions”

In response to generative AI’s rapid development and popularity growth, a group of MTSU faculty from various departments formed a professional development group for the purpose of putting “MTSU faculty at the forefront of the AI revolution,” according to the group’s webpage.

The group, known as the “AI Champions,” hosts meetings, training and workshops to train MTSU faculty on how to use generative AI as professors. The AI Champions have several events scheduled for how to use the technology for the recording industry, history, political science, theater and dance.

“There’s probably not a department or discipline that is gonna be unaffected by some use of generative AI,” said O’Neill, who is a member of the AI Champions. Creative disciplines will be especially affected by the development of generative AI.

“Generative AI creates new opportunities,” said Keith Gamble, director of the university’s Data Science Institute and leader for the AI Champions, in an email to Sidelines. “Workers who use generative AI will replace workers who don’t, just as workers who use personal computers replaced those who could not or would not.”

Part of the champions’ mission is to educate faculty and students about generative AI and to clear up misconceptions about the technology.

“Replacing your work with generated results will not prepare you for in-class tests or to participate in class discussions,” said Wallin, another member of the AI Champions. Using generative AI to produce text without understanding of the content will make “garbage” results.

“AI systems are not ‘magical answer machines,’ they are conversational machines,” he said.

“If you let Generative AI replace your voice, it means you have nothing to say,” Wallin continued. “However, you can often use it as a partner to help take a jumble of thoughts and term them into coherent prose.” 

For the champions, the use of generative AI is not a threat, but an opportunity to equip students for employment in a world that will make use of that technology.

“There’s no choice, you have to embrace it,” O’Neill said. “It is the calculator of the new millennia.”

To contact Editor-in-Chief Matthew Giffin, email [email protected].

To contact News Editor Kailee Shores, Assistant News Editor Alyssa Williams and Assistant News Editor Zoe Naylor, email [email protected].

For more news, visit www.mtsusidelines.com, or follow us on Facebook at MTSU Sidelines or on X and Instagram at @mtsusidelines.

Related articles

‘Sidelines’ 2024: A year in review

Feature photo by MTSU Story by Bailey Brantingham 2024 has been a nonstop year for the MTSU community. From a snowstorm...

Campus workers met by faculty while marching on campus for bi-weekly pay

Featured photo by Maia O'Brien Story by Noah McLane, Kameron Scott and Maia O’Brien MTSU campus workers marched on campus...

MTSU in negotiations with Sila Development to bring a hotel to campus 

Featured photo courtesy of Sila Development Story by Noah McLane MTSU leadership wants to bring a $26 million hotel and...

What is the MTSU CARE Team?

Story and featured photo by Deasia Powell The MTSU Campus Action Response and Engagement team works to assess reports...