Allegations of AI Utilization Surface for High School Junior
Ailsa Ostovitz, a 17-year-old student at Eleanor Roosevelt High School in Maryland, faces accusations of employing artificial intelligence (AI) across three assignments spanning two classes this academic year.
Her reflection on the situation encapsulates a sense of exhaustion: “It’s mentally exhausting because it’s like I know this is my work. I know that this is my brain putting words and concepts onto paper for other people to comprehend,” she articulates.
In a recent communication with NPR, Ostovitz disclosed one particular allegation made by a teacher in September. The message she received contained a screenshot from an AI detection tool, suggesting a 30.76% likelihood that she had used AI for a writing assignment detailing her musical interests.
“I write about music. I love music. Why would I use AI to craft something that I genuinely enjoy discussing?” she questioned fervently.
In an attempt to address the issue, Ostovitz reached out to her teacher using the school’s online platform, expressing, “I said, seriously, I didn’t use AI. Can you try a different detector?” However, her plea went unanswered, resulting in a diminished grade.
Stephanie Rizk, Ostovitz’s mother, voiced her concerns over the allegations, emphasizing her daughter’s dedication to academics and her dismay regarding the premature judgment cast by the teacher. “Get to know their level of skill, and then maybe your AI detector is useful,” she advised.
Rizk later met with the teacher in mid-November, only to learn that the educator had not seen Ostovitz’s earlier message.
District Statements on Detection Tools
Prince George’s County Public Schools clarified in a statement that the teacher employed the AI detection tool independently, as the district does not finance such software.
The district noted, “During staff training, we advise educators not to rely on such tools, as multiple sources have documented their potential inaccuracies and inconsistencies.”
The district opted not to make Ostovitz’s teacher available for discussion. Following their meeting, Rizk mentioned that the teacher no longer believed that Ostovitz used AI.
AI Detection Tools on the Rise
Notably, more than 40% of teachers, ranging from 6th to 12th grade, incorporated AI detection tools in the previous academic year, as reported by a nationally representative poll conducted by the Center for Democracy and Technology.
This trend continues despite numerous studies indicating significant shortcomings in the reliability of these detection mechanisms.
Mike Perkins, a prominent researcher in academic integrity and AI at British University Vietnam, pointed out, “It’s now fairly well established in the academic integrity field that these tools are not fit for purpose.”
Analysis has demonstrated that widely-used detectors, such as Turnitin and GPTZero, have frequently misidentified human writing as AI-generated.
Despite recognized flaws, school districts across states such as Utah, Ohio, and Alabama continue to invest substantial sums in these tools.
Case Study: Broward County’s Approach
Broward County Public Schools, located near Miami, has committed over $550,000 to a three-year contract with Turnitin, which recently expanded its capabilities to include AI detection.
This software produces a percentage indicating the likelihood that a student’s submission was generated by AI; however, a threshold score of 20% or lower is deemed less reliable.
“The Turnitin tool is something that helps us facilitate conversation and feedback, not grading,” remarked Sherri Wilson, director of innovative learning for the Broward district, which serves over 230,000 students. She acknowledged the research on the limits of AI detection accuracy.
Turnitin corroborated this stance, highlighting that its AI writing detection feature may yield inaccurate results and should not exclusively dictate disciplinary actions against students. Wilson emphasized that teachers ultimately retain authority over validating student work.
Usage and Limitations Among Educators
John Grady, a language and literature educator, describes AI detection tools as a means to initiate dialogue with students regarding potential AI usage. “It’s certainly not foolproof,” he asserted. Grady reviews all student essays through GPTZero, considering any assignment with over a 50% likelihood of AI involvement for further scrutiny.
This scrutiny entails examining students’ revision histories, allowing Grady to engage with students whose work raises red flags. He follows up with, “Hey, this flagged. Can you talk to me about why?”
Edward Tian, co-founder and CEO of GPTZero, emphasized that educators should not use the tool as an ultimate verdict but as a part of a broader strategy for student engagement.
Concerns from Students and Educators
Students like Shaker Heights junior Zi Shi express reservations regarding AI detection software. As a non-native English speaker, he worries about the implications of being flagged for possible AI involvement due to vocabulary constraints. “It was definitely disappointing to see the comment being flagged as AI,” he expressed.
Meanwhile, Carrie Cofer, a high school English teacher, scrutinized the efficacy of AI detection systems after conducting her own experiment, in which her Ph.D. dissertation returned a false positive for AI authorship.
Cofer is actively involved in developing her district’s AI policy, advocating against the allocation of funds towards such detection tools.
“I don’t think it’s an efficacious use of their money,” she stated, suggesting that prospective funds be redirected toward teacher training instead.
Adapting to Changing Educational Landscapes
Back in Maryland, Ostovitz has also adjusted her approach, now ensuring that her homework undergoes multiple AI detection checks before submission.
While the writing remains authentically hers, she dedicates additional time to revise sentences flagged by AI detectors, a process that extends her workload significantly.
“I think I’ve definitely become more vigilant about presenting my work as mine and not AI,” she concluded, motivated to safeguard her academic integrity.
Source link: Wgbh.org.
