We can’t help it! We’re always thinking ahead to different possibilities in journal publishing and where the industry is moving.
You too? :)
Last year, members of the Scholastica team came together to share predictions on how the world of scholarly communication would change in 2025, as well as recommendations to help journal publishers and editors keep pace. As 2026 takes off, we wanted to do the same for this year.
Change in scholarly publishing can be sudden (as with the AI boom). It is most often iterative, and above all, it is persistent. The topics we ruminated on this year bear similarities to last but with new developments. Here’s a preview:
- Tracking the impacts of AI on submission volumes and quality (Brian Cody, Scholastica Co-Founder and CEO)
- Keeping up with new accessibility standards and nearing deadlines (Cory Schires, Scholastica Co-Founder and CTO)
- GEO rising and the ongoing importance of metadata in discovery (Danielle Padula, Scholastica Head of Marketing)
- Online fatigue as a catapult for in-person engagement (Katie Sanderson, Scholastica Business Development Specialist)
Tracking the impacts of AI on submission volumes and quality
I’m very interested to watch two overlapping trends in 2026: first, the scale of journal submissions increases as a result of AI; and second, to what degree in our industry LLMs are seen as primarily used to generate AI slop versus contributing to meaningful productivity gains for scholars.
In 2026, will manuscript submissions to journals be 10% higher than last year, or more like 50% higher? We have early indications that some journals/publishers/disciplines are likely seeing submission volume increase in that 50% range (for example, see this discussion of Scientific Reports in Journalology), but I don’t think we yet have an industry-wide sense of whether most publishers should expect to see 5% more submissions in 2026, a number which is likely manageable within existing workflows and team bandwidth, or something more like 50%, which is likely much less manageable for most editorial teams (for more on submissions as drivers of cost, see this Scholarly Kitchen article by Tim Vines).
From my perspective, as an industry, we spent a good bit of 2025 making the (healthy) transition from “we need tools to detect AI because it’s bad” to considering the more thoughtful question: “when and how are LLMs acceptable and when are they not acceptable?” I’m interested in what we’ll learn throughout 2026 regarding how much AI will lead to increased journal submissions generally, and how much of those additional submissions will be unpublishable “AI slop” versus AI-augmented genuine research outputs. My (maybe overly optimistic) hope is that over this and really the next few years, we’ll start to see a decrease in the volume of low-quality AI-generated articles as systems to detect/deter them mature, and that we’ll see broader productivity gains from scholars using LLM-augmented research tools to conduct more high-quality research than they previously could. I also imagine that we’ll see more early-career scholars able to tackle slightly more ambitious research projects than they could have in the past by utilizing LLM tools, which could drive more journal submissions that are hopefully the kind of novel, high-quality content editors want.
So in summary: I’m very curious in 2026 to watch how journal submission volume intersects with LLM usage and article quality!
Keeping up with new accessibility standards and nearing deadlines
In our 2025 scholarly publishing trends post, I talked about the present and future adoption of AI in our industry. And while we have seen some changes (e.g., better models, agents everywhere, and so on), I’m not sure my outlook has significantly changed.
So rather than pontificating about AI, I’d like to talk about a different trend: Accessibility.
In the context of software development, supporting accessibility means developing technology that people of all abilities can perceive, understand, navigate, and interact with effectively. For organizations, accessibility is both a question of inclusivity and legal compliance.
Software accessibility is hardly a new concern. What’s different about accessibility in 2026?
Awareness, adoption, and monitoring of software accessibility have been gradually increasing for decades. But, in 2026, several factors are coming together and will significantly change how we write software:
- Establishment of robust standards (i.e., WCAG outlines clear, verifiable guidelines).
- Greater international harmonization (i.e., widespread adoption of WCAG standards makes international compliance much easier).
- Growing legal clarity and enforcement (i.e., rules and required deadlines are now explicit).
In my opinion, perhaps the most significant driver is growing legal clarity and enforcement. In April 2024, the Department of Justice issued a binding federal rule requiring state and local government websites, apps, and digital content to meet the WCAG 2.1 AA standards by April 2026 (or 2027 for smaller institutions). Similarly, the European Accessibility Act (EAA) now requires both public and many private sector businesses to meet WCAG 2.1 AA standards – including businesses outside the EU (if they offer digital goods or services to EU consumers). Taken together, this means that for those institutions and their software vendors, accessibility rules are about to get much stricter.
We’ve seen a significant increase in accessibility questions and concerns from Scholastica customers. They want to be on the right side of the law and expect – quite understandably – their software vendors to assist them in this goal. As a result, like many other software companies, we’ve made accessibility our top priority. I expect that trend to continue for at least the first half of 2026.
Accessibility compliance brings new challenges. However, in the end, it is for the greater good to make the web more inclusive for all types of scholars. It’s very important work.
GEO rising and the ongoing importance of metadata in discovery
As AI answers become more prominent in Google search results and people turn to GPT, Claude, and other AI tools for research support, there’s little question that Generative Engine Optimization (GEO) will become more critical for publishers.
Of course, there are various aspects of the AI-search future to evaluate in terms of copyright, risks of revenue loss from zero-click AI searches, and new potential monetization opportunities through AI licensing agreements. I expect that publishers will be carefully considering all of those factors over the next year (and beyond). There’s been a lot of thoughtful recent industry analyses on those topics, including this two-part Scholarly Kitchen series on predictions for AI content monetization.
For the purposes of this blog post, I’ll focus on optimizing content for AI searches. This year will be about publishers pinpointing the content they want to prioritize for GEO work (because you can’t do everything at once!) and layering GEO into their existing SEO strategy. I encourage people to look to the low-hanging fruit first, like your publication website and marketing materials.
What does a solid GEO strategy look like? From what we know (despite the elusive nature of AI search), the principles of GEO aren’t so different from SEO. Some of the primary considerations Publishers will have to focus on include:
- Providing clear content summaries whenever possible (for research articles, you can help authors workshop their abstracts toward this goal)
- Leveraging supplementary content that’s more conducive to GEO as a tool to drive people to your research (e.g., a publication blog or podcast in a question and answer format that drills into the punchlines of an article and links to it — this type of content will be more likely to show up in a coveted AI snippet than the article itself for Google search)
- Following good content structure conventions (so clear hierarchies of H1, H2, H3 headings that break up information into short, focused sections that people and AIs can easily parse)
- Tailoring content to user intent (logical placement within your overarching website structure, targeted long-tail keywords, etc.)
- Grouping content on related topics to demonstrate topical authority and semantic depth
In a world increasingly oriented toward AI search, metadata will also become more significant to research discovery than ever (and not just in the traditional sense). Yes, metadata is crucial for optimizing content for indexing in scholarly databases, which researchers will continue to use along with AI, for now at least (a good academic isn’t going to write a paper on AI overviews alone, I hope!). However, metadata also serves another purpose; it functions as a powerful tool to connect related research objects (e.g., code, datasets, article content). As search experiences become more fractured with AI tools, metadata will be the essential glue to connect related information and reveal underlying stories of how research objects came about and how they relate, helping humans and AI verify their provenance.
Online fatigue as a catapult for in-person engagement
From the rise of remote work stemming from the Covid-19 pandemic to the rapid integration of AI into our daily lives, our world has become increasingly virtual. I expect to see rejection of these trends in 2026 and a push towards more in-person engagement opportunities, such as conferences and other meetups. Face-to-face interactions create networking opportunities, collaboration, and connections that virtual spaces have yet to reproduce. While some conferences and meetups have remained virtual due to lingering concerns from the pandemic and hesitancy or inability to travel related to geopolitical disruptions and uncertainties, I think we’ll see high turnouts for in-person conferences and also more local meetups this year, in addition to larger annual events.
I expect many conferences will continue to focus on the twin excitement and anxiety surrounding AI in the scholarly publishing industry and the world at large. Just as live panels and workshops can help educate our community about the role of AI in academia today and tools to help combat some of its challenges, I expect that organic in-person conversations around those topics will be just as useful in navigating the new AI frontier. Face-to-face meet-ups facilitate conversations between people from different organizations who face similar challenges in their own work, enabling them to bounce ideas off each other that they likely wouldn’t have been able to otherwise. I expect to see some commiseration around shared obstacles and, hopefully, also the realization that some concerns might not be as bad as they seem!
Your turn
That’s a wrap on the Scholastica team’s predications and musings on scholarly publishing in 2026 — for today at least! Now, we’d love to hear from you.
What scholarly publishing trends are you watching this year and why? Share your thoughts in the comments section! You can also find Scholastica on LinkedIn and Bluesky.
