Essays written under Standard Chatbot sat closest to AI-only essays in semantic embedding space, and Engage-to-Unlock differed significantly from Standard Chatbot along PC1¶
Claim · All claims
Evidence · 1 study · causalr2·q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person
Subclaims¶
Reading the evidence codes
q is the design tier, i the impact, n the sample, and the kind with its r
the rigour judged within that kind: what each code means.
This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.
q2 i? In exploratory joint PCA of 498 essays, Standard Chatbot had the PC1 centroid (M=−0.014) closest to the AI-only corpus (M=+0.291), and Engage-to-Unlock differed significantly from Standard Chatbot along PC1 (pHolm<.001) but not from Time-Matched Unlock (pHolm=.104). → Xiaotian Su 2026
Evidence¶
Xiaotian Su 2026¶
Xiaotian Su, Laura Rimell, Jiazheng Li, Amal Rannen-Triki, Ulrich Paquet, Lisa Anne Hendricks, Rida Qadri, Daphne Ippolito and Piotr Mirowski. (2026). Who Thinks First? Designing Productive Friction with Engage-to-Unlock GenAI. https://arxiv.org/abs/2610.01518 doi:10.48550/arxiv.2610.01518
q2 · i? · causal · r2
Exploratory joint PCA of participant and AI-only essay embeddings (Gemini Embedding 2, L2-normalized); PC1 explained 17.91% of variance with an unadjusted omnibus F(4, 493)= 255.62, p<.001. The article labels these analyses exploratory and notes dispersion was descriptively similar across human conditions.
"Planned pairwise contrasts (Holm correction,𝑚= 2per component) showed that Engage-to-Unlock differed significantly from Standard Chatbot along PC1 (𝑝Holm < .001), whereas the contrast between Engage-to-Unlock and Time-Matched Unlock was not statistically significant (𝑝Holm =. 104)."
Discussion¶
Related Claims¶
- Engage-to-Unlock participants incorporated less verbatim AI-response text into final essays than Standard Chatbot participants — related
- In this experiment, drafts—human-written or AI-generated—anchored the final essay, with AI-generated drafts revised somewhat more but underlying phrases largely retained across conditions. — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
standard-chatbot-essays-closest-to-ai-only-semantics |
title |
Essays written under Standard Chatbot sat closest to AI-only essays in semantic embedding space, and Engage-to-Unlock differed significantly from Standard Chatbot along PC1 |
description |
Essays written under Standard Chatbot sat closest to AI-only essays in semantic embedding space, and Engage-to-Unlock differed significantly from Standard Chatbot along PC1 |
status |
review |
evidence_strength |
weak |
newest_evidence |
2026 |
generated |
by process:wiki-ingest on 2026-10-10 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — xiaotian-su-2026 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.