"If All We Do Is Catch AI-Created Assignments, What's the Point?"...The Drastically Changing U.S. Education Landscape

Assessment Focus Shifts from Assignment Results to Learning Process
"What Matters More Than AI Use Is Whether Students Are Actually Learning"

As cheating on assignments using generative artificial intelligence (AI) increases, the American education sector is changing its assessment methods. Rather than trying to catch AI use, the focus is shifting to verifying what students are actually learning.

The photo is not related to the specific content of the article. Pixabay

U.S. Education Sector: What Matters Is Not Whether AI Is Used, but Whether Students Are Learning

On September 6 (local time), the Washington Post (WP) reported that methods for dealing with AI-related cheating in American schools are changing. New attempts are underway, such as revising how assignments are assessed and allowing limited use of AI in class.

Rachel Zeleny, director of the writing program at the University of Baltimore, implemented a system last year that recorded the process of students writing and revising essays on Google Docs, after seeing a surge in suspected AI-generated assignments. However, she considered this approach to be little more than temporary surveillance of students and decided to completely overhaul the writing process this year.

Instead of only submitting the final result, students are now required to submit chatbot conversation logs, handwritten outlines, and videos explaining how they completed their assignments. The goal is not to determine whether AI was used, but to verify whether students developed their own ideas and completed their writing directly.

Andrew Keiser, a professor of supply chain management at John Carroll University, uses an online supply chain simulation game. After making important decisions in the game, students are asked to explain the reasons and logic behind those decisions.

Professor Keiser said, "Right now, my biggest concern is not whether students are using AI, but whether students are still learning."

Christopher Ostro, a professor at the University of Colorado, launched a course called "Writing in the Age of AI," where students learn the advantages and disadvantages of generative AI and set their own rules on how much AI they will use for the final project.

AI Utilization. AP Yonhap News

New York and LA Restrict AI Use...AI Companies Expand Services to Teachers and Students

However, responses from educational institutions vary. Recently, New York City temporarily banned the use of generative AI at the high school level, and the Los Angeles Unified School District also temporarily prohibited generative AI use on all school-issued devices.

Meanwhile, AI companies such as OpenAI and Anthropic are offering free services to teachers and school districts. Google is also expanding its presence in education by opening its AI chatbot to some elementary, middle, and high school students.

Concerns About 'AI Agents' Taking Exams in Place of Students...Education Sector Says "Blocking AI Is Not Enough"

There are growing concerns within the education sector that AI agents could log into school systems on behalf of students to take exams or complete assignments.

Anna Mills, an English instructor at Marin College, asserted, "Developers need to restrict AI agent access to school software or implement safeguards to confirm when AI is doing the work in place of students."

However, there is a growing recognition in educational settings that it is difficult to solve these problems merely by blocking AI or relying on detection technology. Professor Ostro pointed out that there is "no technological fix" that can return the learning environment to what it was before the spread of AI.

Hannah Charlin Petty, a high school teacher in Oakland, California, emphasized, "For students to avoid completely delegating their assignments to AI, school must be a worthwhile and genuinely useful place to go."

This content was produced with the assistance of AI translation services.