Summary
According to a post on X by @kimmonismus, a scan of 13,500 arXiv papers from 2021 to 2026 found that 38% of new papers read as AI-written, compared with 0.4% before ChatGPT. The figure comes from a study shared by Chubby, and the post had about 102 likes when the snippet was captured. The result describes how the papers read, not a confirmed count of AI authorship.
AI-assisted summary based on the listed source.
What happened
A scan of 13,500 arXiv papers from 2021 to 2026 found that more than a third (38%) of new papers read as AI-written, against 0.4% before ChatGPT, according to the study shared by Chubby (about 102 likes). Why it matters: Reviewers and journals need new ways to judge research quality when most writing help comes...
Why it matters
If AI now helps with much of the writing in research papers, reviewers and journals may need new ways to judge research quality. Researchers, editors, and anyone relying on writing style as a signal of effort or originality could be affected. Watch for the full study and for any response from arXiv or journals.
Signal Intelligence
Signal Strength 88%
Technical label WATCH
Public Interest 40
Category RESEARCH
Reader Depth TECHNICAL
Signal Strength reflects source quality, relevance, freshness and evidence. Public Interest helps organize discovery; it is not proof of truth.
Public Interest components
Recognizable Entity Score 51
Practical Impact Score 0
Novelty Interest Score 94
Consequence Score 18
Curiosity Score 0
Shareability Score 56
Why this is here
VQV surfaced this signal because it is recent, relevant to Consumer AI, connected to Scobleizer News Pulse (X).