From Static to Dynamic: Rethinking Education in the Age of the Generative Web

Artificial intelligence is already reshaping the way we live, work, and learn. In schools, its presence is felt everywhere: from lesson planning and resource creation to feedback, assessment, and adaptive practice.

For the past three decades, the internet has been built on an archive model: pre-made content, hosted on websites, indexed by search engines, waiting to be retrieved. Education has adapted around this architecture. We teach pupils how to search, sift, and evaluate. We warn them about the risks of unreliable sources, we design assessments that assume access to information, and we spend considerable effort developing their ability to navigate a world of premade content.

But that world is shifting.

---

From Static to Dynamic

Generative AI challenges the current logic of the web. Content no longer has to be there before you ask for it. Instead, it can be conjured up instantly in response to a prompt. Already, studies suggest that the majority of new web pages published this year contain AI-generated material. Some forecasts indicate that up to 90% of online content may be AI-produced by the end of the decade.

This is more than an incremental change. It represents a motion from a static internet, built on fixed, hosted documents, to a dynamic internet, where your query itself creates the content.

• Static model: information exists in advance and can be located, retrieved, and compared.
• Dynamic model: information exists on demand and may be unique to you, generated from the same underlying data but shaped by your prompt, your history, and your context.

The web becomes less a shared library and more a personalised conversation. My version of the internet won’t look like yours. For education, that changes not just what we teach, but how students understand truth, authority, and knowledge itself.

---

What Happens to Human Creativity?

When AI can generate almost any type of content on demand, humans are less likely to produce routine material. Why write “10 tips for revising GCSE English” when an AI can come up with hundreds of versions instantly?

I suspect the incentive to create won’t vanish altogether; but there will be a shift. Human creativity will move into areas where authenticity, credibility, and lived experience matter. In a sea of machine-generated text and AI slop, human voice becomes premium.

For education, this raises a key question: what do we value? Are we still preparing students to produce content that machines now generate more quickly, or are we preparing them to stand out precisely because they are human?

---

From Content Control to System Control

This shift also disrupts how society regulates knowledge. Today’s laws assume content has a source, a host, a publisher. But what happens when the material never existed before it appeared on your screen? Do we regulate each output, or the system that generated it?

I suspect regulators will demand provenance systems, watermarks, and auditing. The risk, however, is that heavy compliance favours the biggest players and squeezes out experimentation and innovation. The danger here will be centralisation: the internet could become less open, less diverse, and more controlled by a few platforms able to shoulder the burden of compliance. Arguably, we are already seing this.

For education, this matters. If access to generative tools becomes unevenly distributed and premium for some, but restricted for others, we risk amplifying rather than narrowing divides.

---

Polarisation in a Dynamic Web

The internet as we have known it at least gives us a common archive to point to. The dynamic internet offers personalised, ephemeral answers that may differ subtly, or significantly, from one person to the next.

This creates the risk of polarisation by design. Two students could ask the same question and receive different narratives, reinforcing different assumptions. Without shared reference points, dialogue itself becomes harder.

In education, that means teaching not only how to question what you see, but also how to understand what others might be seeing. The epistemic humility to recognise that my internet is not the same as yours has already become an essential attribute in the algorithmic feeding of social media, and will likely require strengthening.

---

Evidence-Informed Pedagogy in a Dynamic Web

The move from static to dynamic also challenges how we think about teaching itself, especially through the lens of evidence-informed pedagogy.

• From a more traditional standpoint:
  Evidence-informed teaching often emphasises structured knowledge, memory, and explicit instruction. A static internet reinforces this model: students are encouraged to retrieve reliable knowledge from a fixed archive and commit it to long-term memory to fuel critical thinking. In a dynamic internet, the challenge becomes knowing whether the knowledge retrieved is stable, verified, and cumulative. Traditional pedagogy will likely adapt to a dynamic internet by doubling down on the importance of core knowledge, shared facts, concepts, and methods that can anchor students in a world of shifting answers.
• From a more progressive standpoint:
  Progressive approaches often emphasise inquiry, dialogue, and co-construction of knowledge. A dynamic internet amplifies this ethos, as students can literally be in dialogue with a system that generates knowledge on demand. But it also introduces new risks: inquiry can easily slip into echo chambers if students aren’t taught to question how their generated answers were shaped. Progressive pedagogy will likely adapt by equipping learners with metacognitive strategies to interrogate not just what is generated but why it was generated that way, and how another learner’s answer might differ.
• The common ground:
  Both traditions share a renewed responsibility: to help students develop judgement. Whether through structured retrieval or open-ended inquiry, the emphasis shifts towards teaching learners how to distinguish the durable from the ephimeral, and the authentic from the artificial. In the age of the generative web, evidence-informed practice will need to focus less narrowly on what works best in classrooms in general, and more on how we help students navigate the acquisition of knowledge in a fluid, dynamic environment.

---

Implications for Schools and Universities

So what does this mean for teaching and learning? A few reflections:

Assessment: We will need to design assessments that privilege originality, reflection, and judgement. I suspect hand-writing will therefore remain crucial to assessment, and that oracy will continue to become more important as a way of demonstrating reasoning and understanding.

Information Literacy: Students must learn to interrogate a stream of personalised, ephemeral answers, and to ask not only is this true? but also why am I being shown this version of the truth, and what might others be seeing instead?”

Curriculum: Acquiring secure knowledge remains essential, as it is the foundation for critical and creative thought. But in a dynamic web, the focus must also be on working with knowledge actively: questioning, critiquing, and applying it in unfamiliar contexts.

Ethics & Agency: The key challenge is not just what AI can produce, but who decides what it should produce, under what conditions, and whose values shape those decisions. Students must learn to see content generation as the exercise of power and responsibility, not just convenience.

Community: In an increasingly personalised web, schools may play a new role as one of the few remaining places where common texts, shared discussions, and collective inquiry are guaranteed.

---

A Compass Point

If the static internet gave us the skills of search, the dynamic internet demands the skills of navigation.

For educators, this is both unsettling and liberating. We have the chance to move away from training students to manage archives and towards equipping them to thrive in dialogue with knowledge, with machines, and with each other.

The future of education in the age of the generative web isn’t about fighting the tide of AI-produced material. It’s about preparing young people to live and learn well in a world where information is no longer stored but summoned, no longer fixed but fluid, and, perhaps most challenging of all, no longer shared in the same way by everyone.

---

This builds on themes from the STAR teaching framework and my previous post on what schools should consider when adopting AI, which you might also find helpful.

The shift from a static to a dynamic internet reinforces both points: strategy and pedagogy, not tools alone, will determine how well schools prepare students for this new landscape.

“The irony is that even as digitization is making an increasing amount of information available, it is diminishing the space required for deep, concentrated thought“

— Henry Kissinger

---

• Bluesky
• LinkedIn
• Facebook
• X
• Mail