Since its launch in November 2022, ChatGPT has taken over headlines across higher education media. From admissions to in-classroom, critics are calling out all the ways the AI tool will change the face of education. But while ChatGPT may very well signal a new age of digital assistance, it may not be as drastic as these headlines would have you believe.
What is ChatGPT?
ChatGPT is a predictive text program built by OpenAI. Predictive text programs are Artificial Intelligence (AI)-powered tools that are able to simulate a conversation. Predictive text programs have been around for years, but whereas previous programs were quite formulaic in their responses, ChatGPT is able to produce more fluid and coherent responses that much more closely resemble natural human speech.
How does this affect admissions?
For decades, one of the key components of the higher education admissions process has been the essay. A piece of writing intended to assess an applicant’s motivations and interests, as well as their critical thinking and writing skills, the admissions essay compliments a student's application to create a more holistic view beyond GPA and test scores.
So what are the implications to application fraud if applicants are able to have ChatGPT write a convincing essay on their behalf?
The reality of ChatGPT's impact on admissions
Admissions officers have been defending against various methods of disingenuous applications for years — ChatGPT is simply the newest, and most advanced, iteration.
“The Photomath app, for example, has been around since 2014, with over 300M downloads, allowing students to take a picture of their math problems and get the answers. On the humanities side, papers have been for sale for years,” Dr. Jason Wingard, President of Temple University explained in his interview with Forbes. “Educators and admissions officers are not only well aware of these ‘services,’ but have learned how to spot and work around them.”
Researchers at Forbes decided to test this theory by having ChatGPT write two college essays based on prompts from the most recent Common Application. What they found was that the results, while impressive for a robot, would not ultimately impress an admissions committee.
“I found both essays to resemble cliché essays, with neither answering the prompt in a convincing way,”
shared Jim Jump, the academic dean and director of college counselling at St. Christopher’s School who was asked to review the essays. When asked to weigh in on whether the AI-produced essays would fool an admissions officer, Jump found that, while he probably couldn’t detect the AI authorship, the essay certainly wouldn’t be doing the applicant any favours in the admissions process.
So while ChatGPT isn’t completely up-ending the admissions essay yet, there are some changes you can make in your admissions process to set yourself up for success as this, and similar technologies, evolve and continue to impact higher ed admissions.
Here are four ways you can adjust your admissions process to future-proof it against ChatGPT and other new tech.
4 ways we see higher ed responding to the implications of ChatGPT
1. Incorporating more robust holistic review
Many schools use an admissions essay to gain insight into an applicant’s soft skills and personality traits. Evaluated alongside their GPA and test scores, the essay seeks to provide a qualitative component to the holistic review. But what happens if that essay is fraudulent?
In response, many schools are adjusting their holistic review to include more and more varied evaluation components, reducing the potential for one component of the admissions process, an essay written by ChatGPT for example, to significant impact their enrollment decisions. Asynchronous video assessments, in-person or virtual demonstrated-interest events, and letters of recommendation, are some of the many ways that schools are evaluating applicants’ qualitative skills, interests, and motivations which aren’t as susceptible to AI interference.
This guide on Demystifying Holistic Admissions shares four strategies for incorporating comprehensive holistic review, including case studies from leading schools who have implemented these changes in their admissions processes.
2. Building a responsive process
Technology is constantly evolving. As much as teams would like to stay five steps ahead of any admissions-impacting inventions, the reality is that we’re usually reacting and adjusting to changes as they appear. Some schools, however, are still finding ways to be proactive.
By using third-party vendor platforms instead of in-house systems, schools can adjust or pivot their admissions assessment faster and with less resources. Programs working with Kira, for example, are able to leverage multiple styles of assessment and interviews all within one platform, which is helping them maintain the integrity of their admissions process without compromising the comprehensive review process. Platforms like Kira are also consistently updated to meet industry standards and best practices, helping our partner schools stay up-to-date without adding any work for their teams.
3. Educating applicants on academic integrity, expectations, and consequences
Where the headlines tend to focus on the few bad eggs, in reality most applicants are honourable in their efforts toward gaining acceptance to college. However, as new technologies arise, it might not always be clear to applicants what’s allowed and what’s not.
Several schools, including the University at Buffalo in New York and Furman University in Greenville, SC are addressing this by taking extra measures to educate applicants about how AI tools can and can’t be used in academic settings.
“We want to prevent things from happening rather than catch them when they do,”
Kelly Ahuna, director of the University of Buffalo’s Office of Academic Integrity shared with the New York Times.
In admissions, schools are adding short blurbs or videos at the beginning of the admissions assessment explaining their institution’s standards and expectations for academic integrity, and outlining the consequences for applicants caught not adhering to these standards.
4. Selecting the right tech to defend your admissions process
Although technology might be creating the problem, in this case technology will also offer many of the solutions.
Turnitin, a leading provider of academic integrity solutions, is retooling their robust plagiarism detection software to identify and flag written submissions generated by ChatGPT or other AI authoring tools. OpenAI, the company behind ChatGPT, is also working on an answer to the problems posed by ChatGPT with their launch of a classifier tool which helps detect if AI was used to author text.
“The company is studying hiding cryptographic signals, called watermarks, in ChatGPT results, so that they’ll be more easily identifiable by [anti-cheating software] companies like Turnitin,” Scott Aaronson, a computer science professor at the University of Texas, Austin and guest researcher at OpenAI, shared with Forbes.
Many schools are incorporating these AI and plagiarism detection tools directly into their reviewing processes through enrollment management systems and assessment platform integrations. Turnitin’s SimCheck, for example, is integrated into all assessments in Kira Talent. Running in the background, SimCheck flags suspicious material in admissions written submissions, eliminating the need to rely on reviewers and leaving them free to focus on evaluating the substance of any written content.
Want to know how Kira is helping admissions teams address ChatGPT and other AI authoring tools in our assessments?