Who Should Lead Educational Publishing Evolution: In-House Teams
or External Experts?

I remember a time in the early 2000s when educational publishing felt like a well-oiled machine. If you taught a class or took a course back then, there was usually one dependable source for textbooks: a major publisher with a big team of editors, content specialists, and designers. Textbooks were heavily vetted to match whatever curriculum was in place. This approach made perfect sense for decades.

Then came the shift to digital. And if that wasn’t enough, artificial intelligence (AI) arrived, bringing with it tools that can churn out new material. Suddenly, summaries, question banks, or entire chapters were being produced much faster than any human could.

Teachers can now build lesson plans and tutorials using AI-powered tools. The once-ironclad dominance of publishing houses is giving way to a more fluid and commoditized approach to educational content creation.

Like any industry under siege, the expectation is clear — adapt. Smaller entities leveraging AI are flooding the market with low-cost, passable materials, eroding the competitiveness that once set major publishers apart.

 

How Educational Publishers Are Responding to
AI-Led Content Development

As an answer, according to a recent industry survey, a majority of large education publishers reported using or planning to use AI for adaptive learning content, indicating widespread adoption of AI-driven content development.

Many educational publishers have added generative AI tools to digital textbooks. Pearson, for instance, started rolling out AI features in some of its e-textbooks. By mid-2024, over 70,000 students at 1,000+ institutions were using these AI-enabled textbooks. It can summarize chapters and give students support without them ever needing to leave the textbook.

It’s not just them. McGraw-Hill also rolled out its AI-based platforms  –  the ALEKS assessment system and SmartBook adaptive e-textbooks. Teachers who tested these tools reported higher student engagement and better overall understanding of the material. That’s no small feat!

 

Pressures on Publishing Pricing and Differentiation

With abundance comes price erosion. By automating large portions of content production, AI can reduce development costs (drastically) –  and these savings could be passed on to consumers. A study on AI-generated writing found that if you use AI to draft content (with light human editing) you save up to 90%. That’s a 91% drop in the average cost of content creation​.

UCLA tried this. Their case study showed an AI-developed textbook being offered to students for $25 (versus nearly $200 for previous course materials).

If AI lowers the barrier to entry for content creation, it allows EdTech startups to produce learning content at scale. Publishers are woefully aware that tech companies and open-resource platforms are encroaching on their territory​. Margins are shrinking.

This already played out: the homework-help company Chegg saw a 50% stock plummet in a single day after it revealed that students were turning to free AI tools instead of paying for its services.

Publishers who once had comfortable control over academic content now find that AI is giving newcomers and smaller outfits a real chance to compete with bigger names in the field. They find themselves revisiting the familiar debate of whether to cultivate talent in-house or outsource expertise – this time for AI.

 

A Precarious Tightrope: In-House Teams and Outsourcing

Build AI capacity in-house, or outsource it?

I’ve seen some publishers take steps in early 2023 to build AI solutions internally from the ground up. They took an in-house approach and embedded AI themselves. This strategy gives them control over the pedagogical approach. AI is trained on their style guides and curricula, which in turn, preserves the publisher’s unique voice and rigor. With sufficient investment, a traditional publisher can build AI capabilities that both innovate and uphold content quality, entirely internally.

Publishers with a successful in-house AI strategy take a hybrid approach to content. They combine AI with the expertise of their editorial teams. As an example, AI might generate quiz questions but content developers then curate those questions. This kind of workflow is becoming standard in forward-thinking publishing houses.

It speeds up revision cycles. Your staff can focus on higher-order tasks (like crafting narrative and pedagogy) while AI churns through tedious updates or localization of content for different state standards.

Publishers like Twinkl (a digital educational publisher) publicly state that they use AI to support their teams, not to replace them. Maintaining teacher-approved quality is non-negotiable even as they scale up content production. Such in-house philosophies help ensure that innovation does not come at the cost of trust and accuracy.

But for most, developing AI entirely in-house can be challenging, leading publishers to seek acquisitions or partnerships.

Those who outsource or license AI technology are focusing on strategic partnerships. Some publishers collaborate with AI startups or academic labs to co-develop content solutions. They combine algorithms with the publisher’s content expertise​. This accelerates innovation, but it also means you’ll be sharing control (and revenue) with tech partners.

Ultimately, it often comes down to a publisher’s resources and core competencies.

Larger publishing firms lean towards building or acquiring their own AI solutions to differentiate their products. Smaller publishers might opt for AI solution partners to upgrade offerings quickly. Both models can work, but success lies in content accuracy and editorial oversight, regardless of where the AI originates.

Simply plugging in an off-the-shelf AI generator without customization or oversight is a recipe for mediocre content. Usually, even the output from outsourced AI is managed by in-house subject matter experts.

 

Shift in User Expectations

Analysts project AI in the publishing market to skyrocket from roughly $2.8 billion in 2023 to over $41 billion by 2033, an annual growth rate above 30%.Modern learners expect content to be responsive to their needs. 86% of students globally are already using AI tools in their coursework. They appreciate when a digital textbook can instantly clarify a confusing paragraph or generate practice problems on a concept they struggle with.

In short, the market is signaling: “We want AI-enhanced content, but we need it to be trustworthy, ethical, and complement human teaching rather than undermine it.” User experience research shows that students like AI features like step-by-step solution guides or personalized quizzes but not gimmicky features. I’ve personally seen that the most successful AI implementations have been those tightly aligned with user needs.

 

The Best Deal in Educational Publishing

Market forces are volatile, and those who fail to adapt quickly – or who rely on outdated models – may be left behind by competitors using AI and by consumers flocking to cheaper or more interactive alternatives. Adapting to the demands of speed, innovation, and personalization, needs a strategic approach.

For projects that need deep knowledge of your brand’s vision and long-term strategy, in-house teams are essential.

It’s possible to innovate from within but at the same time, openness to partnerships and new talent can accelerate the AI learning curve. No publisher operates in a vacuum, and collaboration with tech experts can broaden your capabilities​.

My take? Combine the speed of outsourcing with the innovation of in-house teams. Publishers that strike the right balance. Couple AI’s capabilities with prudent guardrails and human judgment and you won’t lose the trust you’ve built over decades. Success will belong to those who are agile but responsible.