Locust Walk with students in fall, University of Pennsylvania, University City area, Philadelphia, … [+]
In the world of college admissions, the AI revolution is here. The use of AI in college admissions is multifaceted. Chatbots, such as those powered by Element451 and Salesforce, are now a common presence on college websites, providing instant responses to applicant queries and serving as virtual advisors. But AI’s influence extends beyond these automated conversationalists. More sophisticated AI applications are beginning to undertake the meticulous task of evaluating transcripts, parsing through AP courses, and getting into the bones of the student’s application, including extracurricular activities and essays.
Regarding AI’s role in evaluating application materials, Professor of Communication at Ithaca College and Higher Education Advisor, Diane Gayeski. says that AI “can look at the number of extracurriculars. It can look know whether you’re a captain of your team or the president of the honor society. The technology can take the rubrics given to an admissions reader and give them to AI.” Gayeski’s testimony suggests that initial screenings of applicants’ transcripts, essays, extracurricular activities and overall profile information can be scanned and coded by AI rather than human readers.
In addition to evaluating application materials, it is expected that colleges will use AI to make final admissions decisions, and a relatively small sample suggests that this practice is well underway. A study conducted by intelligent.com shows that of the 223 responding institutions, 87% use AI to make final admissions decisions. The sample of 223 institutions included 156 higher education institutions, according to the organization, and of these institutions, 43% “sometimes” use AI to make the final admissions decision, and 44% say that they “always” use AI to make the final admissions decision. The majority of these institutions (119) are considered “large.”
In evaluating and selecting applicants, the potential for AI to contribute to bias rather than mitigate it is a concern echoed by educational institutions. University of Texas at Austin experimented with using AI to support evaluation of its Computer Science PhD candidates. They developed an AI admissions screener, only to abandon it upon realizing that it might reinforce existing stereotypes. Georgia Tech is experimenting with using AI in admissions processes given the increase in applicants, but hasn’t implemented it as of May 2023. These examples underscore the need for vigilance and continuous evaluation as AI is integrated into admissions processes.
In using AI to make admissions decisions, colleges may also grapple with the transparency that AI demands. There have been pushes by college admissions experts and analysts to increase admissions transparency, but few data about admissions criteria are known except when institutions provide them through the Common Data Set. The test-optional movement has made college admissions even more opaque, but AI could swing the pendulum back towards clarity—if applied judiciously. At the same time, if AI’s criteria and weighting become public, there’s a risk that applicants will try to game the system. Yet, this transparency is a double-edged sword; it could also lead to a more equitable admissions landscape, where students from all backgrounds understand what’s valued and why.
As students are applying with AI-driven application systems, it is important that they continue to chart their own course rather than change themselves to fit AI-imposed criteria. However, it is important to keep apprised of admissions criteria in order to effectively demonstrate interest, and present their story and experiences to their targeted colleges.
The ethical implications of AI in decision-making cannot be ignored. While AI can efficiently sort applications into preliminary categories, the nuanced judgment required for final decisions will be context-specific based on institutional priorities. For more selective programs, where the line between acceptance and rejection is fine, the subjective nature of human evaluation is irreplaceable. It’s critical that colleges articulate the role of AI in making admissions decisions, ensuring that it complements rather than replaces human judgment.
Credit: Source link
