By: Lucaiah Smith-Miodownik, News Writer
The Peak recently spoke with Simon Fraser Student Society (SFSS) student advocacy coordinator Trish Everett on how SFU has approached the issue of inappropriate artificial intelligence (AI) usage in classes.
As advocacy coordinator, Everett helps students maneuver withdrawals from extenuating circumstances and through academic integrity hearings. These occur when a student has been accused of academic misconduct or dishonesty. “My role is to start on the student’s side and help them do what’s in their best interest,” said Everett. “Oftentimes that’s helping them understand how to navigate through existing processes.”
SFU states that while AI “can be used constructively to support learning,” it may also “interfere with meeting the learning objectives” in courses. Learning objectives include the ability to “analyze and interpret ideas” or “evaluate and use source materials.” AI use is a relatively new facet of academic dishonesty and misconduct. As a result, policies around the usage of AI are still being developed at SFU. “There’s no overarching university policy,” said Everett.
Under SFU’s Student Academic Integrity Policy, academic dishonesty and misconduct both refer to actions that violate “the standard of academic integrity,” which can include the dishonest use of AI. After an accusation (likely from “an instructor, department chair, or the registrar,”) the student will meet with the accuser, who will either assign a penalty or drop the case. Most commonly, a first-time offence yields “a failing grade for the assignment” in question.
If a student asserts they were falsely accused of academic dishonesty or misconduct, the case moves to the University Board of Student Discipline (UBSD) and a hearing takes place. If overturned from here, the Board will “decide upon an appropriate course of action,” and the student has three weeks to create a response for the SFU President, should they decide to. After the President’s decision, if the student still does not accept the outcome, they may be able to appeal to the Senate Committee on Disciplinary Appeals (SCODA). Ultimately, SCODA can rule in favour or against the student based on factual error, excessive penalty, or new evidence.
Everett highlighted some less-recognized subtleties, speaking on how penalties for international and domestic students may vary: “On this particular issue, I would encourage you to think about what would happen if you were suspended as a domestic student, as opposed to what would happen if you were suspended as an international student, and the expenses involved as well as the immigration implications.” She said having to leave the country as an international student could mean breaking a lease to find somewhere to store belongings and paying a significant amount of money to retake classes in the future.
From September 2022 to August 2023, 142 cases of plagiarism were recorded. While the Annual Report on Academic Student Discipline does not divulge the specifics of each case, it does include some examples, such as AI use. Additionally, the report included 370 cases of cheating, including the usage of ChatGPT as an example. The report for the previous year, September 2021 to August 2022, indicates 179 cases of plagiarism and 355 cases of cheating in that period of time.
“To me, it comes down to the question of equity versus equality,” Everett continued. “Currently we have a system that leans towards equal punishment for all, regardless of circumstance. I think an interesting conversation and one well worth having is, is that how we should be doing things, and what roles does equity play when we are addressing [ . . . ] a broad range of different circumstances,” she said.
The type of misconduct a student commits is another nuance Everett discussed. On their third offence, suspension starts to “come into play.” However, the content of these three offences may differ greatly. Everett explained that while one student could fail three final exams by being caught cheating, another student could have committed more “defensible” offences like unknowingly using Grammarly or working with another student against the rules.
“Theoretically, UBSD and SCODA are given leeway to consider the severity,” said Everett. “But again, severity can also be subjective.”
Regardless of international status or the type of offence, rules surrounding AI usage at SFU are unclear. “The most important thing with AI is it’s really changed the landscape of academic integrity and every professor can and does do something different.” She spoke to how “you may have one professor in one course who says absolutely no AI of any kind at all, including Grammarly, and then an hour later you may have another class where your professor is encouraging you to use AI for brainstorming and creating the structure of a paper [ . . . ] and the university is letting all of this happen.”
On May 23, SFU Senate passed a motion stating that SFU will “develop policy regarding the use of generative artificial intelligence by instructors and students in teaching and learning at SFU.”
The Peak reached out to SFU for an update on this. “SFU recognizes the significance of the artificial intelligence (AI) revolution and is working on an AI strategy to determine how we can responsibly leverage and utilize the use of AI for the greater good of education, research, and society,” they responded. They explained their newly-formed AI committee will “look at how the university can explore the opportunities and impacts of AI at SFU in four key areas: learning and teaching, research, operations/administration and external engagement. Task forces and sub-committees are currently being assembled to identify priorities and deliverables in each of these areas.”
They also added, “On September 25, the first townhall event took place, focusing AI in learning and teaching. Several more events are planned — the latest information on the AI strategy and currently scheduled events is available on the AI strategy webpage.”
When it comes to inappropriate AI use, Everett said,“There isn’t a lot of precedent to rely on because there aren’t a lot of AI cases that have gone all the way through the appeals process.” She recognizes that “students don’t commit academic dishonesty for funsies” and that it’s “typically an act of desperation” or “there’s just not time to do the work you need to do,” she said.
SFU students have been vocal about their experience with AI on the SFU Reddit page. One issue raised was professors relying on AI checkers to scan student assignments, which would mistakenly flag their work as being generated by AI. Many students spoke on the burden of having to prove they did not use AI to their accuser.
Everett’s advice? Students should reach out to their professors and let them know when they’re having a tough time. “I think that’s been the biggest take-home lesson of pretty much every case I’ve touched in the last three, four years — if the student had been open with their instructor about the struggle that they were going through, it would have done a lot of good in minimizing the later fallout.”
If you wish to contact Trish Everett, you can reach her by email at [email protected].
