“Empathy, evidently, existed only within the human community, whereas intelligence to some degree could be found throughout every phylum and order including the arachnida.”
― Philip K. Dick, Do Androids Dream of Electric Sheep?
The Brain—is wider than the Sky—
For—put them side by side—
The one the other will contain
With ease—and you—beside—
The Brain is deeper than the sea—
For—hold them—Blue to Blue—
The one the other will absorb—
As sponges—Buckets—do—
The Brain is just the weight of God—
For—Heft them—Pound for Pound—
And they will differ—if they do—
As Syllable from Sound—
~ Emily Dickinson, c. 1862
For better or for worse it’s a “brave new world” in college admissions now that ChatGPT and other generative AI programs are widely available to use both for shaping students’ college applications and for evaluating them to make admission decisions. There are many ethical and practical questions to consider in terms of the varied roles of AI in our lives, so for the purposes of this article I will focus mainly on implications and possible repercussions of harnessing these powerful systems in undergraduate college admissions. Given how rapidly the landscape of higher education is shifting with the employment of AI in classrooms and admission practices, I expect this topic will be the focus of future Spotlight articles as well.
Earlier this month the College Guidance Network, a free online resource for counselors, hosted "Counselor Bites: How AI is Changing Admission" featuring experts Jenny Rickard, President & CEO of the Common App, and Benjamin Lira, Doctoral Student/Researcher at the University of Pennsylvania School of Arts & Sciences (click here to view the recording and related resources). In a nutshell, the webinar examined a number of key questions that pertain to what AI does and doesn’t do well in its current forms:
-
Where does one draw the line between applicants’ original work, especially as it pertains to their college essays, and assistance from either AI or human resources?
-
How much external/artificial support with college applications is appropriate?
-
Can ChatGPT make the college admission process more equitable by providing access to resources and ideas for those who don’t have expert guidance/tools readily available?
-
How might the Common App Essay Prompts and college specific or supplemental essays change to become more “AI proof” in coming admissions cycles?
-
Will more colleges require interviews or “on demand” writing samples or video statements to determine applicants’ authentic voices, presence, and verbal abilities?
-
How can we best harness AI to augment human evaluations, minimize bias, and enhance connections in the application review and college admissions process?
Following from this last question, researchers are now investigating whether and how AI computer models can be trained to read college essays effectively and identify within them an applicant’s character and prosocial traits, thereby scoring “soft” or non-cognitive skills like perseverance, teamwork, and leadership on a scale from 1 to 10. This begs other questions about the limits of both artificial intelligence and human judgment and the lines between human and machine, lived experience and simulation. Are chatbots more or less reliable than people at assessing human qualities, college readiness, and potential in an application? Will those applying to the most selective schools with applicant pools of over 50,000 students need to craft their essays as savvy technographers, brave new journalists seeking to advance beyond initial AI filters to the human interface? While many factors are considered in college admission decisions (see table below), the gatekeepers at the most selective institutions are already using these new technologies in their campaigns to admit the best applicants possible, especially with post-pandemic staff shortages and enrollment management budget constraints:
In my twenty plus years as an educator and guidance counselor, I have been an ardent advocate for student development, humanity, and voice, underscoring their intellectual curiosity and passions in my rec letters, collaborating with them on all aspects of their college apps, including personal statements, and supporting their final decision making for best matches. Considering the possibilities and limitations of large language models (LLM) and AI, it may be difficult or even impossible to interpret accurately through algorithmic “lenses” students’ true identities, directions, and purposes. That said, I expect most of them will “collaborate” with AI programs for essay ideas and structures, to make their personal statements more robust and specific rather than merely cutting and pasting an AI generated response into their Common App essay field, which is likely to backfire or sound a lot like others similarly crafted. And, at the same time, more colleges may ask more specific supplemental questions and/or require additional layers of authentication to avoid generic/boilerplate responses from applicants.
Sage Ridge is in the process of developing a policy statement on the use of generative AI programs, weighing issues of academic integrity and the importance of all our school pillars in student growth and success. To that end I am collaborating with my counterparts at other independent schools and reviewing syllabus statements from colleges across the country, many of which are shared in this growing document: Classroom Policies for AI Generative Tools. On a related note from the National Association of Independent Schools, “Member school teachers reported mixed feelings about artificial intelligence technologies in the latest NAIS Snapshot Survey, including what teachers think about using AI tools like ChatGPT to assist in their teaching and administrative duties, as well as how they feel about students using such tools.”
Interestingly, NAIS is partnering with the Association of Technology Leaders in Independent Schools (ATLIS) on two events to help independent schools learn more about AI in education: (1) the ATLIS and NAIS Webinar: AI in Education (March 30, 1:00 PM ET, virtual; free to NAIS members); and (2) the ATLIS and NAIS Summit: Educational Leadership in the Age of AI (May 3, 8:00 AM-1:00 PM ET, Grand Rapids, MI. Webinar and summit participants will receive an e-book for heads of school and faculty members that provides practical insights on AI in education. Other related resources include:
-
New View EDU podcast: AI and the Future of Education
-
Independent Ideas blog: Navigating the New Normal: Teaching in the Age of AI
-
OPINION: We can add ChatGPT to the latest list of concerns about student cheating, but let’s go deeper (from the founders of Challenge Success)
In the spirit of creativity and to engage OpenAI's newest product, I asked ChatGPT to “Write an article on how AI is reshaping college admissions.” In less than 30 seconds, here in the italicized text is the chatbot’s response:
Artificial intelligence (AI) is revolutionizing every aspect of our lives, and the college admissions process is no exception. With the growing use of AI in the admissions process, universities can now make more informed decisions about who to admit, streamline the application process, and reduce human bias. In this article, we will explore how AI is reshaping college admissions.
-
Personalized Recommendations [for building a college list]
AI can analyze a student's academic history and recommend colleges that match their interests and academic qualifications. For example, College Raptor uses AI to help students find the right college by analyzing their academic and extracurricular achievements, financial situation, and preferences. The tool provides a list of colleges that match the student's criteria, including the likelihood of admission and estimated cost.
-
Predictive Analytics
AI can analyze data from past admissions cycles to identify patterns and predict which students are most likely to be admitted. This analysis can help universities better allocate resources, including financial aid and scholarships, to attract the most promising candidates. For example, Georgia State University used predictive analytics to identify students at risk of dropping out, resulting in a 21% increase in graduation rates.
-
Application Screening
AI can help universities screen applications more efficiently by automating the initial review process. This can save admissions officers a considerable amount of time and reduce the risk of human bias. For example, the University of Edinburgh used AI to screen applications for its computer science program, reducing the workload of admissions staff by 90%.
-
Essay Scoring
AI can grade essays more accurately and efficiently than human readers, enabling universities to evaluate writing skills on a large scale. This can help admissions officers identify candidates who demonstrate critical thinking, communication skills, and creativity. For example, the Educational Testing Service uses AI to grade essays for the Graduate Record Examination, providing a more objective and consistent assessment of candidates' writing skills.
-
Interviewing
AI can conduct virtual interviews and analyze candidates' facial expressions, tone of voice, and language to assess their communication skills and emotional intelligence. This can help universities identify candidates who are a good fit for their programs and culture. For example, the University of Southern California uses an AI-powered interview tool called Kira Talent to evaluate candidates' communication skills and assess their fit with the university.
While AI has the potential to transform the college admissions process, it is important to note that it is not a panacea. AI systems are only as unbiased and accurate as the data they are trained on, and there is always a risk of perpetuating existing biases in the data. Therefore, universities must ensure that AI systems are designed and implemented ethically, and that they are regularly audited for bias and fairness.
In conclusion, AI is reshaping college admissions by providing personalized recommendations, predictive analytics, application screening, essay scoring, and interviewing. By leveraging AI, universities can make more informed decisions about who to admit, streamline the application process, and reduce human bias. While AI is not a silver bullet, it has the potential to make the college admissions process more efficient, effective, and fair.
Reflecting on the impacts that various technologies have had on human culture and civilization over the last few centuries, a few words of caution from Neil Postman seem especially apt: "Anyone who has studied the history of technology [from the printing press to the personal computer] knows that technological change is always a Faustian bargain: Technology giveth and technology taketh away, and not always in equal measure. A new technology sometimes creates more than it destroys. Sometimes it destroys more than it creates. But it is never one-sided.” Neil Postman was a professor of media ecology at NYU and wrote the book Technopoly: The Surrender of Culture to Technology (1992).
