Bibliographic record
Abstract
This article explores the impact of artificial intelligence (AI) on written compositions in education. The study examines participants’ accuracy in distinguishing between texts generated by humans and those produced by generative AI (GenAI). The study challenges the assumption that the listed author of a paper is the one who wrote it, which has implications for formal educational systems. If GenAI text becomes indistinguishable from human-generated text to a human instructor, marker, or grader, it raises concerns about the authenticity of submitted work. This is particularly relevant in post-secondary education, where academic papers are crucial in assessing students’ learning, application, and reflection. The study had 135 participants who were randomly presented with two passages in one session. The passages were on the topic of “How will technology change education?” and were placed into one of three pools based on the source of origin: written by researchers, generated by AI, and searched and copied from the internet. The study found that participants were able to identify human-generated texts with an accuracy rate of 63%. But with an accuracy of only 24% when the composition was AI-generated. However, the study also had limitations, such as limited sample size and an older predecessor of the current GenAI software. Overall, this study highlights the potential impact of AI on education and the need for further research to evaluate comparisons between AI-generated and human-generated text.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.009 | 0.049 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.006 | 0.005 |
| Scholarly communication | 0.006 | 0.005 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.021 | 0.007 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".