The evidence

Everything we argue for on this side of Deskpad rests on more than our own conviction, and it seems only fair to show the work underneath it. These are studies and articles from the wider field, none of them ours, that shaped where we landed on how AI should and should not enter a classroom. We have grouped them by the question each one speaks to, and every card still links straight to the original source, so you can read it yourself and decide whether you see it the way we do.

When AI does the thinking for you

Your Brain on ChatGPT

Accumulation of Cognitive Debt when Using an AI Assistant on Essay Writing Task

This study explores the neural and behavioral consequences of LLM-assisted essay writing. Participants were divided into three groups: LLM, Search Engine, and Brain-only (no tools). Each completed four sessions. A total of 72 participants took part.

Does ChatGPT harm critical thinking abilities?

A new study from researchers at MIT's Media Lab has returned some concerning results.

The study divided 54 subjects (18 to 39 year-olds from the Boston area) into three groups, and asked them to write several SAT essays using OpenAI's ChatGPT, Google's search engine, and nothing at all, respectively.

GPT detectors are biased against non-native English writers

Liang et al., Patterns (Cell Press), 2023.

Researchers ran 91 TOEFL essays written by non-native English speakers through seven widely used AI detectors. On average, 61.3% were wrongly flagged as AI-generated, against about 5.1% for essays by native English writers, a gap the authors trace to the simpler word choices these tools tend to read as machine-like.

Evaluating the accuracy and reliability of AI content detectors in academic contexts

Peer-reviewed, International Journal for Educational Integrity (Springer), 2026.

A recent peer-reviewed evaluation tested today's leading detectors, including Turnitin, under real academic conditions. It found their judgments inconsistent and dependent on context, and concluded that they are not reliable enough to be the sole basis for an academic-misconduct decision and need to be paired with human judgment.

Guidance on AI detection, and why we're disabling Turnitin's AI detector

Vanderbilt University, 2023.

After months of testing Turnitin's AI detector, Vanderbilt turned it off. They noted that a 1% false-positive rate across the roughly 75,000 papers they handle in a year would mean about 750 students wrongly flagged, and concluded that AI detection is not an effective tool that should be used.

New Data Reveal How Many Students Are Using AI to Cheat

But how prevalent is it? New data from a plagiarism-detection company sheds light.

Of more than 200 million writing assignments reviewed by Turnitin's AI detection tool over the past year, some AI use was detected in 1 out of 10 assignments, while only 3 out of every 100 assignments were generated mostly by AI.

AI Gaining Favors in Education with the U-Turn in New York

NYC District Goes From Banning ChatGPT to Exploring AI's Potential.

New York reversed its decision to block ChatGPT on school networks. And now, the district wants to take the lead on crafting policy around the smart use of AI for teaching and learning and the management of schools.

Can AI transform education?

What learning looks like in the future and how to ensure students benefit equally.

Generative AI technology is mind-blowing, with great potential to be harnessed for teaching and learning. In a world where seven out of 10 children born in low- and middle-income countries cannot read by age 10, AI could help address the dramatic learning equity gaps. But this work needs careful thought.

AI in Schools: Pros and Cons

Teachers and administrators need to know AI's advantages and challenges.

Generative AI's impact in education will only grow as AI continues to develop. While the specter of students using AI to cheat has raised concerns among educators, the positive and productive uses of AI counter those fears.