All of this happened quietly. In early 2023, generative artificial intelligence (AI) and chatbots based on large language models (LLMs) were still considered harmless; they were capable only of drafting overly formal text or answering fact-based questions. Over time, however, these chatbots gained new capabilities. First, they learned to address subjective questions; then, they gained access to the internet, enabling them to provide real-time sports scores or live stock market trends. By May 2024, the situation took a sharp turn—a development with significant implications, particularly for the owners of content-driven websites and news publishers.
Google, a key player in the AI race, introduced “AI Overviews” in its search results. This feature presents a small window where Gemini generates a concise summary based on a user’s query. It quickly became a more convenient way to find information, eliminating the need to click through to external websites. Yet, AI Overviews were not an isolated case. OpenAI’s ChatGPT followed suit, integrating an AI-powered search engine into its chatbot. Perplexity developed a native AI-driven answer engine, while Anthropic’s Claude gained the ability to process user-uploaded files and answer questions based on their content. A snowball effect took hold, spreading rapidly.
Although most people may not realize it, a paradigm shift is underway in how the internet is used. Instead of manually searching the web, people are now asking artificial intelligence to curate information for them. This shift, however, is putting pressure on content websites and news publishers.
Today, a significant portion of the content visible on the internet—that is, the indexable websites and pages that appear in search results from Google, Bing, Yahoo, or DuckDuckGo—consists of content-driven sites. These include news platforms (such as the one where you are currently reading this article), social media sites, blogs, forums, and more.
Operated by companies or individual creators, these platforms continuously update their offerings with fresh content, new interactive tools, and enhanced user experiences, all designed to draw in visitors. The ultimate goal of these efforts is to attract a large volume of visitors—what is commonly referred to as “web traffic.” This traffic helps improve a site’s search engine rankings and generates revenue through on-page advertisements.
Until now, the rules have been straightforward: if a website’s content succeeds in attracting a large number of users, it draws in even more visitors, thereby increasing its online visibility and boosting revenue through displayed advertisements. In short, content dictates the flow of internet traffic, and that traffic, in turn, drives revenue generation. Although this system may appear purely transactional, these rules have historically ensured that the quality of content and information remains a top priority. However, with the rise of artificial intelligence, these established rules are beginning to unravel.
A typical AI chatbot with web-search capabilities uses bots—also known as crawlers—to scan hundreds of websites and their content in a matter of seconds. When a user submits a query, the chatbot gathers data, generates a natural-language response, and presents it to the user. While these chatbots often include inline citations pointing to the original sources, how often do you actually click those links to visit the source website when you’ve already received the answer directly within the chat?
So, what happens if only artificial intelligence and its web crawlers visit these content-based websites? For starters, these crawlers are not counted as genuine traffic, as they leave no visible trace. They do not engage with advertisements, and consequently, they generate no revenue. As a result, website traffic declines, and revenue falls along with it.
The Real Impact on Content and News Publishers
The consequences of this shift are already becoming apparent. According to data from Similarweb (as reported by The Economist), global search traffic had fallen by 15% year-on-year as of June 2025. Earlier data also indicates that the “zero-click” rate for news websites—instances where users do not click through to the site—rose from 56% in May 2024 (when “AI Overviews” were first launched) to nearly 69% by May 2025.
For those unfamiliar with the term, “zero-click searches” occur when users find the answers to their search queries directly on the search engine results page (SERP), without needing to click on any of the listed websites. Compounding the issue, Similarweb has observed a decline in organic traffic—dropping from over 2.3 billion visits in the first half of 2024 to fewer than 1.7 billion in June.
This is more than just a matter of numbers. It represents the first fundamental restructuring of the internet since the advent of search engines in the 1990s—a shift for which the world remains largely unprepared. In May of this year, the New York-based financial news outlet Business Insider reportedly laid off 21% of its workforce, citing a need to “respond to extreme declines in traffic that are beyond our control.”
Business Insider is not an isolated case. According to data from Similarweb (cited by the New York Post), traffic to The Huffington Post and The Washington Post plummeted by as much as 50% between April 2022 and April 2025.
The emergence of new technologies is also transforming how internet users access information. According to the Financial Times, 80% of users now encounter AI-generated content in approximately 40% of their search queries.
Problem Statement
At the heart of this entire restructuring lies a fundamental issue. Since the inception of search engines, they have relied on “bots” (crawlers) to scrape and analyze website content, ranking the most relevant pages at the top of search results. This mechanism has historically benefited both search engines and website owners: sites that follow sound content practices earn top rankings, while search engines—by consistently delivering high-quality results—ensure that users keep coming back.
However, AI companies are now employing these same methods to scrape and analyze website content—a process that offers no benefit to publishers. Furthermore, websites are often unable to block these crawlers, as AI platforms frequently deploy them without adequate disclosure. For instance, Reddit recently sued Anthropic after discovering that the latter’s bots had accessed the social media platform as many as 100,000 times in less than a year. The situation with Google is even more complex: the tech giant uses the same bots to scrape content for both its AI systems and its search engine index. Consequently, if a publisher chooses to opt out of having their data analyzed by AI, they effectively vanish from Google Search—a move that essentially kills their traffic.
The Backlash Has Begun
News publishers are well aware of this situation, and they have no intention of remaining silent. The lawsuit filed by Reddit against Anthropic is just one of many similar cases currently winding their way through the courts.
In February of this year, a group of media organizations—including Condé Nast, The Atlantic, Forbes, and The Guardian—filed a lawsuit against the AI startup Cohere, alleging the unauthorized use of copyrighted content from over 4,000 articles. Ziff Davis, the publisher behind sites such as ZDNet, PCMag, CNET, and IGN, has recently filed a lawsuit against OpenAI, alleging “willful and persistent” copyright infringement. In India, the news agency ANI has initiated a similar legal challenge against OpenAI, the developer of ChatGPT.
Meanwhile, the News/Media Alliance—a U.S.-based non-profit industry association representing outlets including The New York Times, The Washington Post, and Vox Media—has issued a statement opposing Google’s AI models, characterizing their operational methods as “theft.” As most of these cases remain pending and no court rulings have yet been issued, no legal precedents have been established. AI companies argue that their use of publicly available information constitutes “transformative use” and does not violate copyright law, whereas publishers contend that accessing content without authorization is, in itself, unlawful.
A Troubling Outlook
While the debate over whether superintelligent AI systems might one day rule the world remains purely hypothetical, the impact of AI on internet search traffic and publishers is very real. Declining website traffic, lower click-through rates, and widespread revenue losses have become the norm, with layoffs increasingly commonplace.
Regardless of the current state of affairs, the wave of artificial intelligence is here to stay. Unless policymakers begin to re-examine content copyright through a modern lens—and courts redefine the concept of “transformative use”—news and content publishers face an uphill battle.
Solutions do exist. Publishers could shift their revenue models away from traditional search engine optimization (SEO) and advertising-driven approaches toward subscription-based models. They might also focus on “AI optimization” to improve their standing within chatbot algorithms, or explore unconventional avenues such as micropayments, newsletters, AI-licensing agreements, or even a “Software as a Service” (SaaS) model.
If the rules of the internet are being rewritten, publishers must adapt to keep pace. Yet, this transition will likely lead to further layoffs and the closure of newsrooms—a reality for which the industry may not yet be prepared.