MétaCan
Menu
Back to cohort

Artificial Intelligence, Academic Misconduct, and the Borg: Why GPT-3 Text Generation in the Higher Education Classroom is Becoming Scary

2023· article· en· W4387122633 on OpenAlexaffvenue
Thomas Mcllwraith, Elizabeth Finnis, Sarah L. Jones

Bibliographic record

VenueAnthropologica · 2023
Typearticle
Languageen
FieldMedicine
TopicArtificial Intelligence in Healthcare and Education
Canadian institutionsUniversity of Guelph
Fundersnot available
KeywordsWonderMisconductPoint (geometry)PsychologyAddictionMathematics educationInterface (matter)Social psychologyComputer scienceLawPolitical scienceMathematics

Abstract

fetched live from OpenAlex

We explore playfully the capacity of an artificial intelligence text generation engine called GPT-3 to produce credible academic texts. Departing from a concern raised by colleagues about the possibility of using GPT-3 to cheat in academia, particularly at the undergraduate level, we interact with the GPT-3 interface as nerdy novices to learn what it could produce. The outputs from the GPT-3 text generation engine are incredible, at times surprising, and often terrible. We point to ways in which GPT-3 might be used by students to produce written work and reasons why most instructors, most of the time, could see through what GPT-3 has produced (at least for now). In our experiments, we learn that GPT-3 can be a productive collaborator in paper design but wonder if this is ethical. In short, while fun and somewhat addictive to experiment with, we must pay attention to the potential ways that AI text generation may begin to appear in the anthropology classroom.

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 imitation

Not 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.

metaresearch head score (Codex)0.045
metaresearch head score (Gemma)0.168
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesMetaresearch, Research integrity
Consensus categoriesnone
DomainCandidate signal: Evaluation · Consensus signal: none
Study designCandidate signal: Theoretical or conceptual · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.994
Threshold uncertainty score0.240

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0450.168
Meta-epidemiology (narrow)0.0000.001
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0010.001
Science and technology studies0.0090.015
Scholarly communication0.0160.019
Open science0.0020.009
Research integrity0.0060.008
Insufficient payload (model declined to judge)0.0180.006

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.

Opus teacher head0.360
GPT teacher head0.479
Teacher spread0.119 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

Study designTheoretical or conceptual
DomainEvaluation
GenreEmpirical

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".

Quick stats

Citations6
Published2023
Admission routes2
Has abstractyes

Explore more

Same venueAnthropologicaSame topicArtificial Intelligence in Healthcare and EducationFrench-language works237,207