DeGenTWeb: When the Internet Becomes an Echo Chamber of AI Talking to Itself
> Paper: DeGenTWeb: A First Look at LLM-dominant Websites > Authors: Sichang Steven He, Calvin Ardi, Ramesh Govindan, Harsha V. Madhyastha > arXiv: 2605.00087 | 2026-05-01
---
1. A New Era of the "Content Farm"
You're searching for a solution to a technical problem. Google recommends a blog that looks professional. The article is well-structured, grammatically flawless, and rich in detail.
But you notice something odd:
- The "latest version" it mentions is actually two years old
- Some technical details sound plausible but don't hold up under scrutiny
- The author has no social media account, no GitHub, no traceable identity at all
- Most of its content is generated by LLMs
- The goal is usually SEO (search engine optimization) — capturing traffic and ad revenue
- Often disguised as human-written blogs, news, reviews, or Q&A posts
- These sites are growing rapidly
- They occupy top positions in some search query results
- Their content quality varies widely — some are nearly indistinguishable from human writing, while others are full of factual errors
What you're reading may not be human-written. It may be LLM-generated.
---
2. LLMs Are "Occupying" the Internet
This paper is the first systematic study of "LLM-dominant websites."
What is an LLM-dominant website?
Key findings:
---
3. Why Is Detection So Hard?
What's wrong with existing LLM-text detectors?
1. High false-positive trade-offs: To avoid flagging human content as AI-generated, detectors are tuned loosely — so they miss a lot of AI content 2. Adversarial evolution: Content farms quickly learn to "bypass" detectors (fine-tuning, post-processing, mixing human and AI content) 3. Domain mismatch: A detector trained on news may fail entirely on technical blogs 4. Language bias: Detectors are less accurate on non-English content
This is an arms race — and the detectors are losing.
---
4. Impact: The Degradation of the Information Ecosystem
What happens when LLM-generated content floods the internet?
1. Training-data pollution: Future LLMs will train on vast amounts of AI-generated text, leading to "model collapse" 2. Search-result devaluation: Finding genuinely valuable human content becomes harder and harder 3. Trust crisis: Readers cannot judge whether content is reliable 4. Distorted economic incentives: Human creators are squeezed out of the market because AI content costs nearly nothing
This is not just a matter of technological progress. It is an information-ecosystem problem.
---
5. A Feynman-Style Judgment: Telling "Real" from "Real-Looking" Is an Old Human Problem
Feynman said:
> "Knowing the name of something" and "knowing something" are completely different.
That's exactly the danger of LLM-generated content: it "looks like" knowledge, but it isn't. It is a probabilistic model recombining statistical patterns from its training data.
When a website is full of content that "looks like" answers — but with no real understanding, no experience, and no accountability — we are losing the ability to distinguish "real" from "real-looking."
---
6. Takeaways
As users and builders of the internet, ask yourself:
1. "How do I verify whether an information source is reliable?" 2. "How much of the online content I consume might be AI-generated?" 3. "If training data is polluted by AI-generated content, what will the next generation of AI look like?" 4. "Do we need new 'digital signatures' to prove the human origin of content?"
DeGenTWeb is a warning. The value of the internet lies not in the quantity of content, but in its authenticity and diversity.
When AI starts generating training data for AI, we are heading toward an "echo chamber" — not an echo of humanity, but a machine talking to itself.