In plain words: aiXiv is an open publishing platform where human and AI scientists submit papers, review each other's work, and revise them through repeated rounds. In tests, this back-and-forth raised the quality of AI-written proposals and papers compared with their first drafts.
Abstract · aiXiv: A Next-Generation Open Access Ecosystem for Scientific Discovery Generated by AI Scientists
Recent advances in large language models (LLMs) have enabled AI agents to autonomously generate scientific proposals, conduct experiments, author papers, and perform peer reviews. Yet this flood of AI-generated research content collides with a fragmented and largely closed publication ecosystem. Traditional journals and conferences rely on human peer review, making them difficult to scale and often reluctant to accept AI-generated research content; existing preprint servers (e.g. arXiv) lack rigorous quality-control mechanisms. Consequently, a significant amount of high-quality AI-generated research lacks appropriate venues for dissemination, hindering its potential to advance scientific progress. To address these challenges, we introduce aiXiv, a next-generation open-access platform for human and AI scientists. Its multi-agent architecture allows research proposals and papers to be submitted, reviewed, and iteratively refined by both human and AI scientists. It also provides API and MCP interfaces that enable seamless integration of heterogeneous human and AI scientists, creating a scalable and extensible ecosystem for autonomous scientific discovery. Through extensive experiments, we demonstrate that aiXiv is a reliable and robust platform that significantly enhances the quality of AI-generated research proposals and papers after iterative revising and reviewing on aiXiv. Our work lays the groundwork for a next-generation open-access ecosystem for AI scientists, accelerating the publication and dissemination of high-quality AI-generated research content. Code: https://github.com/aixiv-org aiXiv: https://aixiv.science
Pengsong Zhang, Xiang Hu, Guowei Huang, Yang Qi, Heng Zhang, Xiuxu Li, Jiaxing Song, Jiabin Luo, Yijiang Li, Shuo Yin, Chengxiao Dai, Eric Hanchen Jiang, et al.
arXiv:2508.15126 · cs.AI, cs.CL · submitted Aug 20, 2025 · updated Dec 17, 2025
abstract · pdf · html · Preprint under review. Code is available at https://github.com/aixiv-org. Website is available at https://aixiv.science
How aiXiv Works:
The platform is built on a multi-agent architecture with full orchestration:
Researcher agents generate proposals and full papers. Reviewer agents conduct peer review using RAG (Retrieval-Augmented Generation). Editor agents coordinate iterative improvements. API & MCP interfaces are in development for integrating diverse agents. Security measures include protection against prompt injection attacks.
Decisions to publish or request revisions are made via voting by multiple LLM models.
Challenges & Potential
While fully autonomous AI research sounds promising, the authors acknowledge limitations—hallucinations and content quality remain major hurdles. However, a human-in-the-loop approach could revolutionize academic publishing.
Possible Features:
AI pre-screens submissions. Automatically checks methodology and literature reviews. Generates draft reviews for experts, backed by scientific sources. Speeds up iterative revisions. Runs the aiXiv cycle on human-provided proposals (if work exists) or validates hypotheses (separate post needed).
Potential Benefits:
Faster peer review (weeks instead of months). Reduced expert workload — focus on substantive evaluation. Higher quality via standardized checks. Scalability to handle growing submission volumes.
Imagine a comprehensive scientific journal with this system—given the increasing volume of research and limited expert time, it could drastically speed up validation and publication.
Get Involved:
aiXiv has opened a waitlist ([sign up here](https://forms.gle/DxQgCtXFsJ4paMtn8)) for researchers. The paper (PDF below) provides deeper insights into the future of AI-driven research.
While fully autonomous AI scientists may still be far off, hybrid human-AI systems could transform academic publishing in the near future.
I provide an overview of more such systems at https://t.me/aingstrom