Well the paywall kept me from commenting based upon the brief part I read. But I did take the title of the article to be a tad alarmist and hyperbolic. DID the article recommend steps to answer the question it posed?
Anyway here's perplexity.ai's take on it.
AI is killing the web. Can anything save it?
The rise of ChatGPT and its rivals is undermining the economic bargain of the internet
Is AI Killing the Web?
The rapid rise of AI tools like ChatGPT and other large language models (LLMs) is dramatically disrupting the fabric of the internet. The foundation of the modern web—the classic exchange where users visit publisher websites in return for quality content supported by advertising—faces existential challenges from AI-driven technologies.
How AI Is Disrupting the Web’s Economic Bargain
AI-driven search and summarization tools increasingly bypass traditional content creators:
Traffic collapse for publishers: Websites, especially content publishers and bloggers, are reporting dramatic traffic losses, with some seeing 20–60% drops in visitors. One travel blog recounted a 90% traffic decline during the rollout of Google’s Gemini AI and eventually shut down after 16 years online.
Bots outnumber humans: For the first time in 2024, bots—mainly AI content scrapers—generated more than 51% of all web traffic. The "crawl-to-visit" ratio (bot activity vs. real human readers) has soared from 2:1 for search engines to over 250:1 for AI crawlers.
Advertising model upended: As users get answers directly from AI and rarely click through to original websites, the core economic model funding free content—the ad-supported web—looks increasingly bleak.
Feedback loop risk: If publishers can’t monetize content, they may cut back or even withdraw from the open web, leading to less new content for AI models to learn from—a “doom cycle” that could degrade the usefulness and creativity of AI itself.
Can Anything Save the Web Economy?
Despite the sweeping changes, several proposals and early-stage solutions aim to adapt the web economy for an AI-dominated era:
1. Content Access Controls
Restricting bot access: Publishers are experimenting with tighter controls on bot traffic using tools like robots.txt and Web Application Firewall (WAF) rules to limit unauthorized scraping by AI crawlers.
“LLMs.txt” and meta-data for AIs: By publishing AI-friendly metadata or new standards (like an “LLMs.txt” file), publishers can better signal how and if their content should be accessed or indexed by AI models.
2. Direct Monetization via APIs
Content licensing and “Cost Per Crawl” (CPCr): Media companies and content creators may shift to paid API models, where LLMs and AI agents must pay to access, summarize, or ingest high-quality content.
Platforms making deals: Some media outlets are already striking licensing agreements with AI providers to ensure compensation.
3. Paywalls and Subscription Models
Making content scarce: As open access becomes less viable, more publishers may move behind paywalls or offer premium content, forcing AIs and users alike to pay for access.
Challenges: However, widespread paywalls risk making quality information less accessible and could accelerate web fragmentation.
4. Building LLM-Friendly Discovery Systems
APIs with attribution and monetization built in: Open standards for AI consumption of content could ensure attribution and payments flow back to original creators, preserving incentives for high-quality publication.
5. Government and Policy Intervention
Regulation and copyright enforcement: Policymakers may introduce laws requiring fair compensation and attribution for AI use of copyrighted content, aiming to rebalance the value equation.
The Web’s Uncertain Future
If nothing changes, traditional web incentives for open, high-quality content could erode, leaving AIs—ironically—starved of the raw material they need to function. There is no silver bullet, but solutions blending content access controls, monetization mechanisms, and potentially regulation, could redefine the relationship between AI, publishers, and users in the internet’s next era.
The question is no longer whether AI will change the web—it already has. The challenge now is inventing models that allow both the open internet and artificial intelligence to evolve together, without destroying the engines of creativity and shared knowledge that made the web thrive in the first place.
AI
Market Data 
Markets 
