- The Midas Report
- Posts
- AI Search Is Quietly Reshaping the Web’s Traffic Economy
AI Search Is Quietly Reshaping the Web’s Traffic Economy
5 minute read
Midas AI Research
July 03, 2025
Search is no longer just a gateway to the internet. It’s becoming a filter… and for publishers, a choke point.
As Google and other search engines roll out AI powered “Overview” results, traditional content providers are watching a foundational source of traffic quietly erode. The shift is happening fast and with significant downstream consequences for digital publishers, advertisers, and the broader content economy.
AI Overviews Are Skipping the Click
In 2024, Google began rolling out its new AI-generated search responses, often labeled “AI Overview.” These snippets synthesize information from across the web and present it directly in the results page… eliminating the need for users to click through to the original sources. While the technology promises speed and convenience for users, it's having the opposite effect on traffic for those who create the content it summarizes.
Multiple major publishers have already reported sharp declines in organic search referrals. Data from SEO platforms like Similarweb and Semrush suggests that AI assisted search flows can reduce click-through rates by double digits on high volume queries. With more than 90% of global search traffic still passing through Google, even small percentage drops translate into major losses in visibility and revenue.
The Ad Dollars Are Moving, Too
The impact isn’t limited to traffic. Advertising budgets are starting to follow users’ changing habits. Research firm MAGNA Global projects that up to $25 billion in ad spend could shift toward AI enhanced search placements and chatbot integrations by 2029. That includes native ad formats embedded directly into AI responses, affiliate placements surfaced through conversational queries, and interactive brand content designed to be pulled into AI summaries.
For many publishers reliant on programmatic ads and branded content traffic, this isn’t just a monetization shift… it’s an existential risk.
Legal and Strategic Responses Are Heating Up
In response, publishers are experimenting with several strategies, some public and some behind closed doors.
Several large media firms are pursuing legal action against companies using their content to train language models without permission. The New York Times filed a landmark copyright lawsuit in late 2023 against OpenAI and Microsoft, and others are expected to follow.
At the same time, forward thinking players are quietly exploring licensing deals with AI companies or creating content that is structured specifically for machine readability. That includes building out content APIs for agent to agent communication and restructuring headlines, summaries, and schema markup to increase visibility in AI generated results.
A New Content Strategy Is Emerging
The shift in user behavior demands more than SEO tweaks… it calls for a new paradigm in content design. Publishers and content creators who want to maintain discoverability need to optimize not just for human readers, but for AI systems scanning for authoritative, digestible, and linkable sources.
That means:
Structuring content for clear attribution and inclusion in AI responses
Using schema.org markup and snippet-friendly summaries
Creating short, standalone answer blocks that AI tools can reference directly
In essence, the battle has moved from search rankings to search visibility in AI summaries.