AI chatbots would weaken students' writing, he warned, and not just grammar and spelling.
"It also destroys the humanity and individuality of writing," Gough said, worrying chatbots would also alter how students interact by encouraging them to rely heavily on algorithms to express themselves.
ChatGPT relies on technology known as a large language model, which is trained to mimic how people write by processing a large database of digitized books and online writings and analyzing how words are sequenced together
That's why Gough recommended the district strengthen polices on academic integrity.
"You would be surprised by the amount of essays and assignments done completely with ChatGPT," he told the board on March 9.
As of this moment, the district lacks a fleshed-out policy on chatbots, but that's not surprising considering everybody is racing to catch up with the new technology.
AI seemingly advanced at warp-speeds this year when platforms like Chat GPT seized the public imagination for its astounding ability to write essays with a convincingly human voice — and alarmed graders of English papers nationwide.
STATE GUIDANCE?
Locally, AI chatbots have sparked conversations among K-12 teachers, university faculty and administrators on how to integrate or curb its use, though there's been little guidance from the state.
The Oregon School Board Association currently has no AI policy on the books, and neither does the Oregon Department of Education.
However, the agency is working on resources to support school districts in crafting their own equity-centered AI polices, according to ODE spokesperson Marc Siegel.
"Cheating and plagiarism are valid concerns that ODE continues to hear from educators and families across grade levels," Siegel wrote in a statement, suggesting districts — in order to be proactive — draft strong AI policies, employ cheat-checking software, and train teachers on its use.
"It is critical that any new tools are used to assist students in their learning, not to avoid learning," he wrote.
So, with the school year fast approaching, how are districts in the Mid-Valley region responding to the emergence of crafty chatbots?
K-12 SCHOOL DISTRICTS
The Corvallis School District's technology services department has had ongoing conversations with cybersecurity consultants around AI, according to district spokesperson Kelly Locey.
The school district hopes to craft an AI policy this fall, she said.
Additionally, Locey said the district would be employing the cheat-checking software Turnitin (think "turn it in") this year. That tool, which detects plagiarism in student assignments, can also spot AI-generated content.
It's also one of the tools Gough recommended the district use back in March when he spoke before the Corvallis school board.
According to Turnitin's AI webpage, however, the company's AI detection software isn't completely foolproof and shouldn't be the only measure of a student's potential tinkering with a chatbot.
Greater Albany Public Schools also currently lacks policy on the use of AI chatbots in the classroom and software to detect it, according to district spokesperson Michelle Steinhebel.
There have been staff-level conversations around its use, though, she said.
The Lebanon Community School District also has had staff-level conversations around AI, according to district spokesperson Susanne Stefani.
While there's currently no district policy on the matter, LCSD is already using Turnitin and is working on updating the district's handbook for teachers and staff, Stefani said.
She likened catching up with AI to trying to jump on an already fast-spinning merry-go-round.
"It's moving so quickly, so we're kind of already a step behind," she said.
OREGON STATE UNIVERSITY
When AI began making its photo-realistic waves earlier this year, the OSU's Center for Teaching and Learning, along with other partners, spearheaded a taskforce to draft guidelines for faculty.
Among other things, they encouraged staff to consider redesigning assignments and updating syllabi on when and how to use AI in the classroom.
Regan Gurung, the CTL's associate vice provost and executive director, updated his own general psychology course to make assignment questions more specific to his class's expressed learning outcomes — a harder task for an AI chatbot to complete.
"If you have a general assignment, ChatGPT is pretty good at giving you a general answer," Gurung said.
Those guidelines, drafted this spring, also don't endorse any specific detection software because none, so far, has proven to be 100 percent accurate.
Instead, Gurung maintains the best tools teachers currently have at their disposal are their eyes.
"Even with smart prompting, ChatGPT still writes like a computer," he said, adding that teachers can pick up on clues in a student's text.
"The downside for us instructors is that we've got to read stuff much more closely," he said.
This upcoming fall, OSU will continue work on drafting a university-wide policy on AI use, and this next year, a new CTL committee on AI will host panels and listening sessions on AI's impact on teaching, research and productivity, Gurung said.
A USEFUL TOOL?
While it remains unclear exactly how AI will impact student writing, Gurung said chatbots will never be a substitute for "knowing how to write."
He compares chatbots like ChatGPT to calculators in math class. You can add sums together on a calculator, but if you don't understand the process of adding, you're still lost.
When Gough shared his concerns with the Corvallis school board this spring, he acknowledged the merits of chat bots in the writing process, saying he used AI himself to brainstorm ideas for essays.
Gurung said chatbots can help students who are stuck on prompts; he thinks it's shortsighted for any educator to completely dismiss the technology's educational benefits.
"The fear that AI will change how students write, will only come true if students solely use AI and turn their brains off," he said.
©2023 Albany Democrat-Herald, Ore. Distributed by Tribune Content Agency, LLC.
