Charting the Future of Scholarly Knowledge with AI: A Community Perspective

A new manuscript calls for greater cross-disciplinary collaboration to build AI systems capable of structuring the world's rapidly expanding body of scholarly knowledge, arguing that isolated research efforts are slowing progress toward integrated solutions [1]. The paper, submitted to arXiv in August 2025 and revised in June 2026, notes that many researchers continue to rely on manual methods for knowledge extraction and organization, often because they are unfamiliar with existing tools or lack access to domain-adapted solutions [1]. The accelerating volume of publications across disciplines has made it increasingly difficult to stay current, the authors write, underscoring the need for scalable, AI-enabled approaches to synthesizing scholarly knowledge [1]. Multiple research communities have independently developed tools and frameworks aimed at building reliable, dynamic, and queryable knowledge bases, but limited interaction among these groups has hindered the exchange of methods, models, and best practices [1]. The manuscript identifies strategies to foster cross-disciplinary dialogue, categorize new collaborations, and shape future research directions [1]. The work emerged from the inaugural AAAI Bridge on AI for Scholarly Communication, held at AAAI 2025, which convened students, domain researchers, practitioners, and AI experts over two days [4]. The event's goal was to identify shared challenges and catalyze collaborations in scholarly knowledge extraction, organization, and use [4]. Discussions at the gathering also addressed ethical considerations, including concerns about bias, provenance, reproducibility, and the responsible use of AI in scholarly contexts [9]. The push for AI-assisted scholarly knowledge organization arrives amid broader structural shifts in scientific publishing. The development of open science has revived debates over linguistic diversity in research, with 120 international organizations signing the 2019 Helsinki Initiative on Multilingualism in Scholarly Communication, which called for supporting infrastructure in national languages [8]. UNESCO's 2021 Recommendation for Open Science included linguistic diversity as a core feature, aiming to make multilingual scientific knowledge openly available and reusable [8]. Economic pressures are also reshaping the landscape. The largest commercial publishers have moved toward author-pay models funded through article processing charges and transformative deals, while diversifying into data analytics and metrics that monitor academic activities [11]. Open science commons — shared ecosystems of journals, platforms, and repositories — have emerged as an alternative, structured around self-governance principles [11]. The manuscript's call for integrated AI solutions arrives at a moment when the infrastructure of scholarly communication is already in flux, with funders and institutions acknowledging that the sector no longer operates under normal market conditions [11].