The Web Is Still Growing, But It’s Not the Same Web Anymore

Over the past two decades, the internet has gone from a publishing space for humans to a machine-readable ecosystem that serves platforms, algorithms, and, now, artificial intelligence. At first, we adapted our writing to meet the demands of search engines. Then, we adapted our content strategies to suit social media feeds. Now, we’re adjusting again—this time to AI-driven search results that summarize information without always sending users to the original source.

This shift has transformed everything: how we write, where content is published, and who controls visibility. Entire categories of creators—from bloggers to newsrooms—have seen their relevance and reach decline, even as content creation in general continues to grow across formats and platforms.

In this post, we’ll explore how we got here and what it means for small creators, independent publishers, and businesses that depend on web visibility:

How SEO-first writing weakened the quality and readability of online content
Why platforms like YouTube reward engagement over truth, and why that’s a problem for journalism
How the web is thriving as a platform for tools and apps, but failing as a home for trusted media
The real-world consequences of AI answering user queries without driving traffic to sources
Why AI still depends on authentic human content—and what happens when that content slows down
And finally, how this entire system is becoming circular, with content feeding AI, and AI reshaping content visibility

We are now at a point where the web still works—but not in the way it used to. And if the current trajectory continues, content creation could collapse under the very systems that claim to support it.

Let’s break it down.

When Writing for Google Broke the Way We Write

For more than a decade, SEO dictated how content was created. Repeating exact-match keywords throughout headings, paragraphs, and image alt tags was considered necessary for ranking. The result was content that sounded unnatural—even unreadable.

This wasn’t just a quirk of digital marketing. It reshaped how language was used online. Many articles became hard to follow, filled with awkward phrases and keyword stuffing designed more for Google’s crawler than for human comprehension.

As a result, clarity took a back seat. Instead of writing for understanding, we were writing for machines. Over time, this started affecting how people expected online content to sound—fragmented, repetitive, and algorithmic.

YouTube Isn’t Journalism—It’s Content Creation

Today, YouTube is one of the most influential platforms for information. But what many call “journalism” on YouTube is often just content made to gain views.

Creators might use headlines that look like news, but their goal is not fact-checking or balanced reporting. It’s engagement. The platform rewards emotion, controversy, and rapid content turnover. This encourages creators to favor virality over truth.

That doesn’t mean all creators are misleading—but the format itself prioritizes attention over accuracy. This is a problem when audiences rely on these videos as primary news sources. It creates an illusion of journalism, without the accountability real journalism demands.

Content Creation Is Expanding, but the Web as a Media Platform Is Shrinking

Despite concerns over declining traffic and content value, there’s no doubt that content creation itself is growing. We’re seeing more formats, more platforms, and more creators than ever. The web feels busier—but not necessarily healthier.

Creators today aren’t just writing blog posts. They’re producing short-form videos, podcasts, infographics, and apps. AI is making it easier to repurpose content across formats. A long-form article can be transformed into a podcast overview, a YouTube script, or a carousel post with minimal effort. This cross-format capability, powered by multimodal AI models, lowers barriers to creation and distribution.

From a technical standpoint, the web as a platform for building tools and apps has never been stronger. Platforms like Figma and even Google itself are now embedding dynamic apps into what used to be static search pages.

But ironically, while content creation and application development are expanding, the web as a platform for media and information is hitting a low point. As noted in a recent conversation with The Verge’s editorial team, if they were starting from scratch today, they wouldn’t even launch a website—they’d go straight to TikTok or YouTube.

This shift has major implications

More content exists, but it’s not being consumed through traditional web channels.
AI is expected to boost content production further, yet most creators struggle to monetize it.
Google says it’s sending more traffic to a wider set of sources, but many publishers report the opposite.

Google claims AI Overviews (its new AI-powered answer inserts) include source links and help users explore more of the web. However, major media alliances argue that Google is taking content without proper compensation. According to them, if traffic were really increasing, publishers wouldn’t be pushing back this hard.

What we’re seeing is a paradox: more content, more formats, and more users—but a weakening of the traditional web model where creators publish, users search, and traffic flows back to original sources.

Google’s AI Inserts Are Cutting Off the Source of Traffic

Google has started including AI-generated responses directly on the search results page (SERP), answering user queries without requiring a click to any website. For users, this might feel convenient. But for site owners, it represents a major loss of traffic.

Matthew Prince (CEO of Cloudflare) confirmed what many have suspected: sites across the web are seeing fewer visitors as more queries are answered within Google itself. If people no longer need to click to get an answer, why would they?

The irony is that the same websites losing traffic are the ones whose content was used to train the AI answering the query. After years of optimizing content for Google, creators now find themselves bypassed by the very system they helped build.

The Coming Content Crisis: AI Still Needs Something to Learn From

There’s an even bigger concern looming: AI models need content to train on. Not just any content, high-quality, current, accurate, human-created material. But what happens when that material dries up?

Many blogs and smaller publishers are cutting back on content creation because there’s no traffic and no return. Meanwhile, major news outlets are moving behind paywalls that block AI crawlers. This means there is less original content available for training future language models.

Some suggest that synthetic content—content written by AI—could be used to fill this gap. But that creates a feedback loop. AI learns from AI, with no new input, and errors compound over time. The result is a drop in quality and reliability across the board.

This could lead to a true content crisis: a situation where there isn’t enough fresh, verifiable material to sustain the tools that now power everything from customer support to search engines.

A System That Eats Itself

What we’re seeing now is a classic case of the snake eating its own tail.

In response to the pressure from ChatGPT and other AI tools, Google is reinventing its search results. Instead of leading users to sources, it delivers answers right on the SERP, no clicks, no attribution, no real incentive to visit the original site.

Why? Because the longer users stay on the search page, the more they’re exposed to Google’s ads. If Google provides a direct answer and links clearly to the source, users leave faster, and Alphabet (Google’s parent company) earns less.

This model has led to tension between Google and content creators. There’s ongoing debate about whether Google should be required to pay publishers for using their content in AI-generated snippets.

So what’s the solution? Some suggest blocking AI bots from crawling their content. But that creates another problem: the content can’t be used to improve AI models either. And without high-quality input, those models stagnate.

This is the same dilemma faced by companies building AI tools. The models they create depend on a steady stream of human content, but the more these tools reduce the need for clicks and traffic, the less incentive creators have to publish anything at all.

It’s a self-defeating cycle that threatens the long-term sustainability of both online content and AI development.

Conclusion: Write for Humans, Structure for AI

The web hasn’t stopped growing, but the way content is found, consumed, and rewarded has changed dramatically. As AI tools take a more central role in how people search for and interact with information, traditional strategies for visibility no longer deliver the same results.

Creating valuable content is still essential. But now, that content must also be structured in ways AI can easily understand, summarize, and cite. This means going beyond old SEO habits. It’s time to think in terms of SEO 2.0—where semantic clarity, contextual relevance, and topic authority are just as important as keywords.

At ai4k.eu, we recommend every creator and business adopt a content strategy that supports both human users and machine understanding.

To help with this, we’ve created

A free checklist for AI visibility to make sure your pages are readable, indexable, and useful to AI models.
Our SEO 2.0 course that teaches how to structure and publish content that stands out in a search environment increasingly powered by AI.

If content is still the fuel of the internet, then strategy is the engine. Don’t get left out of the results just because your website isn’t readable to machines. Write smart. Structure clearly. Stay visible.

At ai4k.eu, we help entrepreneurs, creators, and small businesses build websites and content strategies that focus on long-term value, not short-term tricks. Explore our website plans to see how we help brands build real visibility in an internet where clarity, trust, and consistency still win.

Contact me here.

How AI and Google Are Breaking the Web Content System