Research & Papers

Algoverse AI Research misleads high school students into paying for fraudulent NeurIPS publications

A paid program promises high schoolers NeurIPS publications for $3,325—but the papers are riddled with errors.

Deep Dive

A Reddit post on OpenReview has exposed a concerning program run by Kevin Zhu through his organization Algoverse AI Research. Zhu boasts 158 publications and 468 coauthors, and his company targets high school students with a paid program that claims to help them publish at NeurIPS workshops for a fee of $3,325. Their website states '289 Algoverse Students Accepted to NeurIPS 2025,' implying a high success rate. However, the integrity of these papers is now in question.

When the user randomly examined four papers from the program, each contained glaring errors that should have been caught by reviewers. For example, one paper reported identical numerical results for two different experimental conditions (Stigma Negative vs. Stigma Positive). Others had broken prompts in a dataset claimed to be human-reviewed, results that contradicted the abstract, and AI-generated citations with wrong authors and formats. Additionally, Kevin Zhu added himself as an author on every paper without disclosure. This appears to be a systematic scheme to sell academic credentials to high schoolers, exploiting their desire for elite college admissions.

Key Points
  • Four randomly selected papers from Algoverse all contained fundamental errors, such as identical results in different experimental conditions and AI-generated false citations.
  • Kevin Zhu is listed as an author on every paper, and self-citations are undisclosed; the program charges $3,325 per student for NeurIPS workshop publications.
  • The program claims 289 students accepted to NeurIPS 2025 and markets itself as a shortcut to Stanford and MIT, encouraging academic dishonesty among minors.

Why It Matters

This scheme undermines academic integrity and exploits high school students' college ambitions for profit.