🔒USENIX Security Symposium Sees Record Submissions: +31% Year-Over-Year
AI in Research: The Good and the Hallucinated
TL;DR
The USENIX Security Symposium saw a record-breaking 3,030 submissions this year. However, five reviewers were asked to step down due to suspected misuse of AI in the review process.
USENIX Security Symposium has received an unprecedented number of submissions: 3,030 papers compared to last year's 2,400. The surge is partly attributed to the rise in AI usage among researchers, but it also comes with challenges. Five reviewers were asked to stop participating due to suspected misuse of AI tools during the review process. Of the first-round submissions, 1.78% (21 papers) were rejected for containing three or more hallucinated references. The organizers are drawing a line at using AI for bibliography preparation but allow limited use in polishing human-written text. This year's NDSS saw its submission count jump from 694 to 1,481, reflecting the growing impact of AI on academic research.
Key Points
The USENIX Security Symposium received 3,030 valid submissions this year, up from 2,400 last year
Five reviewers were asked to cease participation due to suspected use of AI during the review process
Of the first-round submissions, 1.78% (21 papers) were rejected for containing three or more hallucinated references
The NDSS saw its submission count jump from 694 in 2024 to 1,481 this year
Conference organizers allow limited use of AI to polish human-written text but ban it for bibliography preparation
Why It Matters
If you're a researcher or reviewer at USENIX Security Symposium, the new guidelines on AI usage will impact your workflow. The rejection rate for papers with hallucinated references is low (1.78%), but the potential consequences of misuse are significant. This shift in policy highlights the need for transparency and integrity in academic research.
Frequently Asked Questions
Why does this matter?
If you're a researcher or reviewer at USENIX Security Symposium, the new guidelines on AI usage will impact your workflow. The rejection rate for papers with hallucinated references is low (1.78%), but the potential consequences of misuse are significant. This shift in policy highlights the need for transparency and integrity in academic research.
What happened?
The USENIX Security Symposium saw a record-breaking 3,030 submissions this year. However, five reviewers were asked to step down due to suspected misuse of AI in the review process.
Comments
Be the first to comment
Enjoyed this article?
Get it daily. 7am. Free. Reads in 5 minutes.
Join 2,680 builders reading daily.