ChatGPT has arrived – and nothing has changed

Last week, a colleague came running into my office in a panic. He had run his final exam questions through ChatGPT, the latest advance in computer language processing, and it had gotten a perfect score (which was much better than most of his students had managed). “This is the end of essays and remote exams!” he told me. “We won’t be able to stop students from cheating!”

The same panic has gripped professors, administrators and policymakers across the sector, as they grapple with the question of ensuring academic integrity in the wake of ChatGPT. But their concerns are neglecting a key fact: we’ve never been able to ensure academic integrity.

Sure, students can now ask ChatGPT to write their essays for them. But students could always hire ghostwriters to write their essays for them. At one university I was affiliated with, there were flyers all around campus openly advertising ghostwriting services offered by PhD students looking to supplement their meagre stipends.

• ChatGPT and the rise of AI writers: how should higher education respond?
• Original essays written in seconds: how ‘transformers’ will change assessment
• Eight ways to engage with AI writers in higher education

Students can now ask ChatGPT for answers to test questions when taking remote exams, but students could always hire others to take remote exams for them. Fraternities and sororities in the US have exam banks and answer keys for previous years’ exams stretching back decades, allowing for easy cheating on tests set by professors who reuse test questions or use assessment materials from textbook companies. Software that prevents computers accessing the web while students are taking an exam can easily be thwarted with a second computer, tablet or phone. 

I’ve even heard apocryphal stories of companies who hire test takers and make-up artists; the make-up artists use their skills to make the professional test takers look enough like the clients to fool ID checks and take in-person tests for desperate students!

Of course, it may be easier and cheaper to use ChatGPT than other ways of getting around academic integrity, but the fact remains that students who are determined to cheat will find ways to do so. Nothing has really changed; academic integrity is difficult to police and always has been.  

If we want to deter cheating, the best way to do so is to remove the motivation for cheating in the first place. That can occur through creating norms of academic integrity. In a classic study by Donald McCabe and his colleagues, the most important contextual determinant of student cheating is the perception of whether other students are cheating. Subsequent research shows that cheating is less prevalent when universities clearly communicate to students that academic integrity is valued and given high priority on campus and when students are given a significant role in adjudicating and upholding an honour code. Students who enjoy a course and respect the professor are also less likely to cheat. Finally, to the extent that learning and credentialing can be divorced (such that cheating no longer provides benefits), students would have no incentives to cheat because there would be no benefit to doing so.

Of course, we could also forestall cheating with ChatGPT by changing assessments and classes such that using Chat GPT isn’t cheating. Once students graduate they will have access to ChatGPT on their phones around the clock; it could be useful to structure classes and assessments to incorporate how one should go about solving problems when one has access to ChatGPT rather than trying to prevent students using it. Indeed, allowing ChatGPT into our assessments would free up cognitive resources for other thinking, much as we’ve done with calculators in many mathematics classes or spelling and grammar checks in many written assignments.

For as long as there has been high-stakes assessment in education, there has been cheating. ChatGPT may make it a little easier for students to cheat, and a little harder for us to catch them if they do, but it doesn’t fundamentally change the integrity dynamics in higher education. The best ways of thwarting cheating have never been focused on policing and enforcement; they have been about integrity training, creating a healthy campus culture and reducing incentives to cheat. There is no need to panic about ChatGPT; instead we can use this as an opportunity to modernise our thinking about academic integrity and ensure we’re using best practices in combating dishonesty in the classroom.

To finish, considering the above, here are some practical ways to prevent cheating in your class:

1. Spend time on the first day of class discussing the importance of academic integrity.
2. Make your expectations about what constitutes cheating clear. If you don’t want students using ChatGPT in your class, tell them so explicitly.
3. Incorporate an honour pledge. Have students sign an agreement that they will not cheat – and have them reaffirm that agreement at the beginning of every assessment during the class.
4. Always treat your students with respect and comport yourself with integrity – be a role model that students can and want to emulate.
5. Use formative rather than summative assessment when possible, such that students are rewarded based on effort rather than achievement.
6. Create assessments that incorporate ChatGPT use, such that using ChatGPT is not cheating. 
    

Danny Oppenheimer is professor jointly appointed in psychology and decision sciences at Carnegie Mellon University. He researches judgement, decision-making, metacognition, learning and causal reasoning, and applies his findings to domains such as charitable giving, consumer behaviour and how to trick students into buying him ice cream.

If you found this interesting and want advice and insight from academics and university staff delivered direct to your inbox each week, sign up for the THE Campus newsletter.