The integration of generative artificial intelligence into search engines has fundamentally altered how web traffic flows, how users interact with information, and how digital marketers measure success. Not all links generated within Google’s AI Overviews are created equal, and recent observations from the search engine optimization (SEO) community highlight a shifting landscape.
As Google continues to refine its AI-driven search experiences—transitioning from static text summaries to dynamic, conversational interfaces—website owners and SEO professionals face a complex web of changing UI elements, ambiguous tracking mechanics, and new data reporting standards in Google Search Console.
This comprehensive analysis examines the mechanics of Google’s AI Overview links, the chronology of recent updates, supporting data from analytics platforms, official guidance from Google executives, and the critical implications this evolution holds for publishers.
Main Facts: Understanding the AI Overview Link Ecosystem
To navigate the current search landscape, publishers must distinguish between the various types of links generated inside Google’s AI Overviews and AI Mode. According to official Google documentation and product announcements, links within generative search results generally fall into several distinct categories:
- Inline Links: Hyperlinks placed directly adjacent to or embedded within the answer text, designed to direct users to relevant external web sources.
- End-of-Response Suggestions: Article recommendations and deeper analyses appended to the bottom of an AI-generated response.
- Discussion Previews: Previews of online community discussions and social media threads, featuring clickable links that display the community or platform name.
- Subscription-Labeled Links: Links integrated into answers that feature specific badges indicating they belong to a user’s active news subscriptions.
- AI Mode Conversation Starters: Follow-up questions or anchor links that, when clicked, do not navigate to an external website; instead, they launch a conversational back-and-forth session within Google’s "AI Mode."
The tracking of these interactions within Google Search Console depends entirely on the destination of the click:
- External Navigation: Clicking a link that leads to an external webpage counts as a standard click.
- Query Refinements: Links or interactive elements that trigger a new Google query (such as launching an AI Mode conversation) typically do not count as a click or an impression for an external site, functioning similarly to query refinement links.
Chronology of Google’s AI Interface Updates (2025–2026)
Google’s rollout of generative search features has been iterative, characterized by continuous interface experimentation and shifting product goals.
- August 2025: Robby Stein, VP of Product at Google Search, outlined the company’s long-term "north star": increasing the visibility and frequency of inline links within AI answers. Desktop link carousels were introduced during this period to encourage deeper web exploration.
- January 2026: Google rolled out global mobile support allowing users to click follow-up questions directly from an AI Overview, instantly transitioning into a fluid, conversational interface via AI Mode.
- February 2026: Stein announced further interface modifications, including desktop pop-up groups for links upon hovering, alongside more descriptive link icons designed to improve user engagement across devices.
- May 6, 2026: Hema Budaraju, VP of Product Management for Search, detailed five major updates to AI Overviews and AI Mode. These included labeled subscription links, discussion forum previews, and expanded inline links with desktop hover previews.
- June 3, 2026: Google formally confirmed an increase in inline links within AI responses, noting ongoing testing of alternative link architectures to drive click-through rates.
- August 31, 2026: Following a regional test in the United Kingdom, Google officially rolled out the dedicated Generative AI performance report globally within Search Console, allowing site owners to filter impressions specifically generated by AI Overviews and AI Mode.
- Late September 2026: Industry analysts captured sudden surges in inline link frequency and documented new anchor-link behaviors routing users straight into AI Mode rather than destination websites.
Supporting Data: What Industry Observations Reveal
Recent data shared by independent search analytics platforms and SEO practitioners sheds light on how these interface updates translate into real-world behavior.
The Rise of External Inline Links
On September 21, 2026, Malte Landwehr of AI search analytics firm Peec AI published a chart illustrating a dramatic shift in Google’s output. Tracking the daily average percentage of AI Overviews containing an external inline link between August 25 and September 20, the data revealed a stagnant baseline near zero for the first two weeks of September.
However, following a sharp inflection point on September 11, the presence of external links within answer text climbed steadily, reaching 26.2% by September 20.
While the dataset lacked broad market breakdowns and query-volume specifics, Landwehr’s observations suggest that Google may be aggressively training users to interact with inline elements, raising questions about whether these links will eventually favor direct web traffic or pivot users toward internal AI Mode queries.
The AI Mode Anchor Link Phenomenon
On the same day, SEO practitioner Gagan Ghotra shared an 18-second video recording demonstrating a newly observed behavior: clicking on bold, underlined anchor text inside an AI Overview did not open an external publisher’s website. Instead, the interaction opened an interactive query thread inside AI Mode.
This finding highlighted an emerging tactic by search engines to keep users within conversational environments. While the exact scale of this test remains unverified by Google, it underscores a growing tension between traditional organic referral traffic and AI-driven conversational journeys.
Official Responses and Search Console Mechanics
Google’s documentation regarding Search Console reporting offers critical clarity—and persistent opacity—for webmasters trying to measure their performance.
The Generative AI Performance Report
The introduction of the Generative AI performance report on August 31, 2026, provided a dedicated mechanism for tracking impressions originating from AI Overviews and AI Mode. This report isolates AI-driven impressions from traditional web search results, allowing publishers to filter data by page, country, date, and device.
However, a fundamental mathematical hurdle remains: the report measures impressions only, not clicks or click-through rates (CTR). Because overall search performance metrics bundle clicks from traditional web listings and AI features together, site owners cannot mathematically isolate a dedicated CTR for AI Overviews. Furthermore, chart totals aggregate impressions at the property level; if multiple pages from the same site appear within a single AI response, they are counted as a single property impression unless filtered by URL.
The Ambiguity of Query Refinements
According to Google Search Console help documentation, links that act as query refinements—executing a new search query rather than visiting an external page—do not register clicks or impressions for third-party properties.
While official guidelines address standard query refinements and AI Mode follow-up questions, the search community lacks explicit documentation regarding whether newly observed in-text anchor links (like those captured in Ghotra’s recording) are classified as standard external links or internal query refinements. If classified as refinements, publishers receiving high visibility in AI Overviews may see zero measurable traffic or Search Console click attribution from those specific interactions.
Implications for Publishers and SEO Professionals
The ongoing transformation of Google’s search engine results page (SERP) carries profound implications for digital publishing, content strategy, and performance measurement:
1. The Traffic Attribution Gap
The inability to track distinct click-through rates from AI Overviews leaves publishers operating in the dark. Even as Search Console displays rising AI impression metrics, content creators cannot definitively link those impressions to downstream revenue or verified site visits.
2. Conversational Intent vs. Direct Referral
As Google encourages conversational exploration via AI Mode, the fundamental user journey is shifting. If search engines successfully transition users from reading external web articles to engaging in multi-turn dialogues with AI assistants, top-of-funnel informational traffic may decline significantly across the web.
3. The Need for Advanced Analytics
Third-party tracking tools and analytics suites will become vital for digital marketers trying to reverse-engineer AI visibility. Relying solely on native Google tools will no longer suffice for understanding true audience acquisition channels.
Looking Ahead
As Google continues to experiment with link carousels, hover previews, subscription badges, and conversational anchor text, the search ecosystem remains in a state of permanent beta.
For publishers and SEO professionals, the immediate path forward requires closely monitoring Search Console’s evolving reporting features, adapting content to satisfy AI citation parameters, and preparing for a future where traditional blue links share structural dominance with generative, AI-mediated pathways. Until Google provides more granular metric attribution for inline AI interactions, vigilance, multi-channel tracking, and agile optimization will remain the primary defenses in an increasingly automated search environment.
