WHO WROTE THIS ESSAY? DETECTING AI-GENERATED WRITING IN SECOND LANGUAGE EDUCATION IN HIGHER EDUCATION

In the citing paper

“A recent study demonstrated this in detailing one such qualitative study comprising a mere six ESL lecturers, highlighting narrow empirical breadth and ongoing needs for broader policy-focused research to inform institutional responses to student work produced via AI [2].”

In the citing paper

“Although an ever-growing volume of literature about AI in education has been generated, much of the research on detection tool research to this point is of limited scope and lacking in institutional breadth [2].”

In the citing paper

“ESL instructors often judged grammatically accurate and sophisticated writing as AI-generated, revealing a bias against second-language writers and highlighting how current detection systems may unintentionally penalize linguistic diversity [2].”

In the citing paper

“Educators rely on subjective judgments without clear policies, sometimes misinterpreting linguistic sophistication as evidence of AI authorship.”

In the citing paper

“Faculty in ESL contexts often feel uncertain about addressing suspected AI-generated student work, highlighting the lack of comprehensive, AI-specific institutional guidelines [2].”

In the citing paper

“The authors noted that all the detection tools produced reports indicating the probability that the content had been generated by AI [2] and highlighted their probabilistic and often inconsistent nature.”

In the citing paper

“ESL instructors often relied on superficial linguistic cues, mistakenly equating fluency with artificiality.”

In the citing paper

“At the same time, AI detection tools showed inconsistent results and lacked transparency in their classification criteria, complicating their role in making academic decisions [2].”

In the citing paper

“Some researchers have studied AI detectors in an ESL context and found notable discrepancies between the judgments made by humans regarding the origin of the content versus what machine detectors indicated.”

In the citing paper

“In their investigation, ESL teachers were only 33% to 66% correct in identifying whether texts were composed by students or produced by AI [2], highlighting the dilemma teachers experience in terms of being able to make credible distinctions.”

In the citing paper

“The lack of standardized procedures for flagged submissions compounds these technical shortcomings [2].”

In the citing paper

“Furthermore, institutions are inconsistent in interpreting AI-assisted work under academic integrity policies [2].”

In the citing paper

“It was found that AI detectors worked more effectively in identifying AI-generated texts than humans, and AI, to some extent, challenged lecturers' previous evaluation criteria and strategies.”

In the citing paper

“It was found that AI detectors worked more effectively in identifying AI-generated texts than humans, and AI, to some extent, challenged lecturers' previous evaluation criteria and strategies.”

In the citing paper

“particularly when many academics have reported insufficient AI literacy (Alexander et al., 2023).”

In the citing paper

“Fifty-four percent of the reviewed articles highlighted that the emergence of GenAI presented a significant challenge to student academic integrity (Adeshola & Adepoju, 2023; Alexander et al., 2023; Chaudhry et al., 2023; Chan, 2023; Dai et al., 2023; Elsayed, 2023; Geerling et al., 2023; Kang, 2023; Nikolic et al., 2023; Ouh et al., 2023; Overono & Ditta, 2023; Perkins, 2023; Smolansky et al., 2”