More GenAI-susceptible courses show no raised grade floor after ChatGPT: passing-margin and failing/withdrawal probabilities are null under both COVID assumptions¶

Claim · All claims
Evidence · 1 study · causal r2 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? The probability of clearing the passing margin (at least a D) and the probability of a failing grade or withdrawal show no significant change in more susceptible courses after ChatGPT, under either COVID modeling assumption. → Dumlao 2026

Evidence¶

Dumlao 2026¶

Dumlao, J. M. Z., Wang, M., Xie, Z., Hu, J., Bar, I., Chaney, G., III, Gold, H., & Teplitskiy, M. (2026). Generative AI Availability, Grades, and Student Satisfaction at a Large University. arXiv. https://arxiv.org/abs/2607.21534 doi:10.48550/arxiv.2607.21534

q2 · i? · causal · r2

Grade-distribution DiD analysis estimating the preferred specification on letter-grade threshold indicators. Both bottom-end outcomes "are both null under both bounds" and pass the parallel pre-trends test (COVID-excluded p = 0.458 and 0.255), which the authors call a credibly identified null.

"The bottom of the distribution is cleaner. The probability of clearing the passing margin (at least a D) and the probability of a failing grade or withdrawal are both null under both bounds."

Discussion¶

Related Claims¶

• The share of students earning at least an A rises in more susceptible courses after ChatGPT under the preferred bound, but the shift fails parallel trends and is read as descriptive, not causal — related
• Positive average grade effects appear only under COVID-laden treatment anchors, indicating such readings reflect the COVID shock rather than ChatGPT — related
• GenAI availability produced no significant differential effect on final grades in more GenAI-susceptible courses relative to less susceptible ones — related
• Effects of GenAI availability on self-reported understanding are insignificant; effects on interest are significant only under the transient-COVID assumption — related
• Grade effects of GenAI availability do not differ across terciles of students' prior academic preparation, robust across multiple ability proxies — related
• Students received significantly higher homework scores during GenAI-evaluation periods, but this advantage did not transfer to downstream exams — related