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 highquality 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.

Introduction. The modern scientific method has long enabled groundbreaking advances in science and technology, but its progress is fundamentally limited by researchers’ ingenuity, background knowledge, and finite time (Lu et al. 2024). For decades, AI researchers have aimed to automate scientific discovery (King et al. 2004; Reddy and Shojaee 2025; Zhang et al. 2025a; Liu, Li, and Wang 2025), starting with early symbolic systems that replicated hypothesis formation and scientific reasoning (Segler, Preuss, and Waller 2018). More recently, the advent of Large Language Models (LLMs) has revolutionized this field (Bai et al. 2023; Touvron et al. 2023; Jiang et al. 2023; Zhang et al. 2025b; Brown et al. 2020), enabling AI agents to autonomously generate scientific proposals (Hu et al. 2024; Si, Yang, and Hashimoto 2024), conduct experiments (Lu et al. 2024; Schmidgall et al. 2025), author papers (Lu et al. 2024; Zou et al. 2025), and per- form peer reviews (Zhu et al. 2025a; Yixuan et al. 2024; Ryan and Nihar 2023).

Discussion / Conclusion. In this work, we presented aiXiv, a next-generation openaccess platform designed to support autonomous scientific research conducted entirely by AI scientists. Unlike traditional journals and preprint servers, aiXiv is built from the ground up to facilitate AI-driven research workflows, enabling agents to autonomously generate, review, and refine scientific content. The platform also offers APIs and MCPs to further facilitate this process. We introduce a closed-loop review system for both proposals and papers, incorporating automatic retrievalaugmented evaluation, reviewer guidance, and robust defenses against prompt injection. Extensive experiments demonstrate that our review-refine pipeline significantly enhances the quality of AI-generated research. Iterative reviews lead to measurable improvements in proposal and paper’s quality.