The AI Discovery Shift: Five Lessons for Publishers

AI is dramatically changing the way we discover and consume scholarly content. As AI-powered search becomes more commonplace, publishers are confronted with both significant challenges and exciting opportunities as they push to ensure their content remains discoverable, well-protected and commercially viable. 

During a recent webinar, which KGL co-hosted alongside our partners at Cashmere, we discussed how publishers can prepare for this rapidly evolving landscape. We also looked at how the KGL AI Gateway can help publishers to establish governed licensing frameworks and explore new monetization opportunities.  

Here are some of the main themes that emerged from our discussion: 

1. AI is changing how scholarly content is discovered

Traditional search behavior, as we once knew it, is changing. Increasingly, researchers are turning to AI-powered tools to answer questions directly and are spending less time clicking through to publisher websites and platforms to find the information they need. 

This noteworthy shift has major implications for publishers, according to José Fossi, VP of Hosting Operations, KGL PubFactory: "It's pretty clear that web traffic is diminishing and that Google's AI overview is preventing access to websites – astonishingly, 68% of Google searches are now zero click searches. This presents a major challenge for vendors and publishers who face the risk of losing control of access attribution, and of course potential monetization via their channels."  

As referral traffic declines, publishers need to find new ways to ensure their content remains visible and discoverable to researchers. This may involve considering AI as another discovery channel like any other; one that requires the same levels of strategic planning and investment as search engines, indexing services and library platforms. 

2. Understanding the difference between AI training and inference

One of the most important topics discussed during the webinar was the difference between training and inference in AI models. 

Training involves using content to build or improve an LLM. Once content has been ingested into a model, it cannot easily be removed. Inference, however, works differently. AI systems retrieve relevant content in real time to generate grounded, accurate responses. This allows publishers to license access to trusted content while retaining greater control over how it can be used. 

With the inference market set to grow by 18.8% year-on-year, Jonathan Munk, Co-founder and CEO at Cashmere, was keen to highlight its potential: "We think the larger commercial opportunity in AI is on the inference side of the market and that's where a lot of growth is about to happen and what we're focusing on here at Cashmere." 

This is an opinion echoed by many industry experts, who believe that inference will become an important long-term, strategic commercial opportunity for scholarly publishers, as they seek to license content in a sustainable way, while maintaining ownership of their IP. 

3. Centralized infrastructure is key to scaling AI opportunities

As AI licensing opportunities continue to grow, managing individual agreements and technical integrations will likely become more complex. The webinar emphasized the benefits of a centralized destination for publishers that can bring together content, rights management and licensing within a single infrastructure. 

This approach, as offered by KGL's AI Gateway, not only allows publishers to manage multiple licensing relationships from one place but also to maintain control over rights and permissions, simplify content delivery and monitor usage and performance across multiple AI applications. Rather than responding to every new AI platform independently, publishers can establish a scalable foundation that supports current and future opportunities. 

4. New licensing models are emerging

While traditional subscription licensing remains important for publishers, they are also increasingly exploring additional approaches. From token-based pricing and institutional AI licenses to collection-specific licensing and "bring your own license" (BYOL) models, AI is creating an array of exploitable new commercial models for the industry. 

The panel went on to discuss best practice approaches towards pricing. Simon Boisseau, VP of Publisher Market Development at KGL Accucoms said: "In most ways, AI licensing is no different from other forms of licensing, it's just delivered in a different way and arguably has more value to the end user and the company who is purchasing it. It is difficult to pinpoint exactly what the right price point is, but we always recommend working with each interested company on an individual basis to understand their use case and to work the pricing strategy out from there." 

5. Experimentation will shape the future

Perhaps the strongest prevailing message from the discussion was that the AI market remains in its early stages. No single business model has emerged as the industry standard, and pricing, licensing structures and technical approaches are all still in the developmental stages. However, the panel was keen to highlight that publishers who begin to experiment now will be better positioned to influence how the market develops in the future. 

"This is a once in a generation shift for this industry and if you aren't in the game, you have no influence on the game. We firmly believe that participation helps you set the rules for how the market is going to work and the more involved with helping us hone how we go about this, the better off the industry will be," concluded Munk.    

AI is transforming scholarly publishing in much the same way that digital publishing and online discovery reshaped the industry over the past two decades.  

While many questions remain, the direction of travel is becoming clearer and publishers have an opportunity, not only to protect their content but also to develop sustainable new revenue streams through AI discovery and licensing.  

By building the infrastructure, partnerships and expertise today, we can all be ready for the opportunities that tomorrow's AI ecosystem will bring. 

Click below to request a link to watch the full webinar, "From Search to AI: The Future of Discovery, Licensing, and Monetization for Publishers" and learn more about the KGL AI Gateway. 

KnowledgeWorks Global Ltd. (KGL) is the industry leader in editorial, production, eLearning, online hosting, and transformative services for every stage of the content lifecycle. We are your source for learning solutions, intelligent automation, research integrity, digital delivery, and more. Email us at info@kwglobal.com.