For nearly three decades, the unspoken contract between Google and the open web was straightforward: publishers provided the high-quality information, and in exchange, Google provided the traffic. It was a symbiotic relationship that fueled the growth of everything from niche hobbyist blogs to global news organizations. But that contract is currently being rewritten in real-time, and many publishers are finding the new terms unacceptable.
The catalyst is the aggressive rollout of AI Overviews (AIO), Google’s generative AI search experience that synthesizes information from across the web into a single, comprehensive answer at the top of the search results page. While this offers a seamless experience for the user, it creates a “zero-click” reality for the creator. When the AI provides the full answer, the incentive for a user to click through to the source website vanishes.
Recent data and reports from publishers, including those highlighted by Heute.at, indicate a significant decline in organic traffic for sites that previously relied on informational queries. In response to this mounting pressure and the evident frustration of the publishing ecosystem, Google is now implementing updates designed to restore a flow of visitors to external websites. The company is attempting a delicate balancing act: keeping the efficiency of AI while avoiding the collapse of the incredibly ecosystem that feeds its models.
The Zero-Click Dilemma and the Publisher’s Plight
The shift toward AI-generated summaries represents a fundamental change in how information is consumed. In traditional search, Google acted as a librarian, pointing users toward the best book on a shelf. With AI Overviews, Google has become the narrator, reading the highlights of the book aloud so the user never has to open the cover.
This “summarization” effect hits different sectors with varying intensity. Transactional queries—such as “buy running shoes”—remain relatively stable because the user still needs to visit a store to complete a purchase. However, informational queries—the bread and butter of news sites, recipe blogs, and tech tutorials—are seeing the steepest declines. If a user asks “How do I remove a wine stain from a rug?” and the AI provides a five-step bulleted list, there is little reason to click on the home-improvement blog that actually conducted the research.
From a technical perspective, this is an issue of Click-Through Rate (CTR). For many publishers, the “featured snippet” was once a trophy that drove traffic. Now, the AI Overview often pushes those snippets further down the page or absorbs them entirely, effectively cannibalizing the organic traffic that sustains ad-supported business models.
Google’s Pivot: Tuning the AI to Save the Web
Google is acutely aware that its AI cannot exist in a vacuum. Large Language Models (LLMs) require a constant stream of fresh, human-generated data to remain accurate and relevant. If publishers go bankrupt or stop publishing because they are no longer receiving traffic, the AI begins to suffer from “model collapse,” where it begins training on its own synthetic data rather than real-world information.
To mitigate this, Google is introducing several adjustments to the AI Overviews experience. These updates are focused on increasing the visibility and “clickability” of the sources used to generate the AI response. Rather than burying links in slight text or hidden menus, the new updates aim to make citations more prominent and integrated into the flow of the answer.
Google is refining the “trigger” mechanism for AI Overviews. Internal adjustments are being made to ensure that AI summaries do not appear for every single query, particularly those where a detailed, long-form article is more valuable than a brief summary. By limiting the frequency of AIOs for certain intent-based searches, Google hopes to preserve the traditional search funnel for high-value content.
| Feature | Traditional Search | AI Overviews (Initial) | Updated AI Overviews |
|---|---|---|---|
| User Action | Clicks link to find answer | Reads summary on SERP | Reads summary, encouraged to click |
| Traffic Flow | High volume to source | Significant “Zero-Click” drop | Attempted recovery of CTR |
| Publisher Value | Ad revenue/Lead gen | Data extraction only | Hybrid attribution model |
| Primary Goal | Navigation/Discovery | Immediate gratification | Balanced utility and ecosystem health |
The Broader Economic Friction
This conflict is not merely a technical glitch but a systemic clash between the old web and the new AI economy. We are seeing a similar pattern emerge with platforms like Perplexity and OpenAI’s SearchGPT, both of which prioritize the answer over the link. This has led to an increase in “robots.txt” blocks, where publishers explicitly tell AI crawlers they are not allowed to index their content.
However, blocking Google is a “nuclear option” for most publishers. Since Google remains the primary gateway to the internet for billions of people, disappearing from the index entirely is often more damaging than losing a percentage of clicks to an AI summary. This gives Google significant leverage in how it decides to “share” the traffic.
The stakes are particularly high for independent journalism and niche expertise. While massive media conglomerates can diversify their revenue through subscriptions or first-party data, small-to-mid-sized publishers rely almost entirely on the visibility provided by search engines. For them, a 20% or 30% drop in organic traffic isn’t just a metric—it’s a threat to their operational viability.
What Remains Uncertain
Despite the updates, several critical questions remain unanswered. First, there is the issue of transparency: Google has not provided a clear formula for how citations are chosen for AI Overviews. Publishers are left guessing whether their content is being surfaced based on authority, keyword density, or some other opaque algorithmic preference.
Second, the effectiveness of these “traffic-boosting” updates is still being debated. While Google claims that AIOs can lead to “more qualified” traffic—meaning the users who do click are more likely to be deeply interested in the topic—publishers argue that “qualified” traffic cannot replace the sheer volume of visitors needed to sustain an ad-based model.
Finally, there is the looming shadow of regulation. In the European Union, the AI Act and existing copyright laws are creating a framework where AI companies may eventually be required to pay for the content they use to train their models or summarize in real-time. This legal pressure may force Google to move beyond mere “UI tweaks” and toward a formal revenue-sharing model with publishers.
The next major checkpoint for this evolution will be Google’s upcoming quarterly earnings reports and subsequent developer briefings, where the company is expected to provide more data on the performance of AI Overviews and their impact on the broader web ecosystem. As the AI landscape shifts, the industry will be watching closely to see if Google can truly coexist with the publishers it relies upon.
Do you think AI summaries are killing the open web, or are they just the next evolution of search? Share your thoughts in the comments below.
Related reading
