My latest observations of the web space confirm a tectonic shift: generative models are rapidly becoming the primary "authors" of the modern internet. Fresh data I managed to analyze shows that roughly 10% of all existing web pages bear clear traces of synthetic generation. However, a far more alarming figure lies in the dynamics: if we consider only content published after ChatGPT's debut in November 2022, the share of AI texts exceeds one-third.
Methodology and Scale of the Phenomenon
During the study, I processed a random sample of 10,000 English-language pages collected via Common Crawl in July 2026. For identification, I used a machine learning model tuned to detect specific linguistic patterns characteristic of generative architectures. It is important to emphasize: these 10% are merely the tip of the iceberg, as the sample included a vast layer of outdated content created long before the era of modern neural networks.
When I filtered out "prehistoric" pages and focused exclusively on materials published after ChatGPT's launch, the picture changed dramatically. More than 33% of new pages demonstrate statistically significant signs of AI involvement. This is direct evidence that synthetic content is growing exponentially alongside the spread of Claude, Gemini, and other competitors.
The Commercial Sector — The Main AI Battleground
The distribution of synthetic content across domain zones is highly uneven. In 2026, every tenth page in the .com zone contains AI generation, which is twice the rate of .org (4.6%). Educational and governmental resources remain the "cleanest" — there, the share does not exceed 1%. This is explainable: commercial platforms are actively automating the production of product descriptions, SEO articles, and news feeds, where speed matters more than uniqueness.
A Critical Caveat on the Data
One should understand the limitations of this estimate. I did not verify the creation history of each page or survey the authors. The methodology relies on probabilistic linguistic markers rather than precise attribution. This concerns content that was either fully written or substantially edited by a neural network. Minor edits using AI assistants do not fall into this category.
My expert conclusion: We are witnessing not just a trend, but a fundamental transformation of the web. The market has already adapted to a new reality where the line between human and machine creativity is blurring. Investors and analysts should take into account that search ranking algorithms and monetization systems will increasingly be restructured to filter synthetic content, creating both risks and new opportunities for projects betting on verified authorship.