The Evolution of Search: Liz Reid Addresses the Future of Google
The landscape of digital information is currently undergoing its most significant transformation since the invention of the search engine itself. As Artificial Intelligence (AI) becomes deeply integrated into the way we find information, questions have arisen regarding the survival of the open web, the relevance of traditional SEO, and the quality of the content we consume. Liz Reid, Google’s Vice President of Search, recently sat down for an in-depth discussion on these very topics, providing a rare glimpse into how the tech giant views the convergence of AI and human-generated content.
In her conversation on a recent Bloomberg podcast, Reid addressed the common anxieties surrounding “AI Overviews,” the shift in how users phrase their queries, and the persistent threat of “AI slop.” Rather than signaling the end of the web, Reid suggests that we are entering an era where AI and websites coexist, creating a more efficient and useful ecosystem for users and creators alike.
Debunking the Myth: Is AI Killing Website Clicks?
One of the most pressing concerns for publishers, bloggers, and SEO professionals is whether AI Overviews—the summarized answers that appear at the top of Google search results—will cannibalize website traffic. If a user can get the answer directly on the Google search results page, why would they ever click through to the source?
Reid’s perspective offers a more nuanced view of user behavior. She distinguishes between “bounce” clicks and “meaningful” engagement. According to Reid, AI Overviews are primarily designed to filter out low-value interactions. These are the instances where a user clicks on a page, quickly scans for a single date, name, or fact, and then immediately hits the back button. These “half-second” visits provide very little value to the publisher and can be frustrating for the user.
By providing those quick facts through AI, Google aims to streamline the search experience. However, Reid emphasizes that for more complex needs—such as reading a long-form analysis, researching a product, or seeking a human perspective—users still want to visit the actual website. “If what you were going to go in and do is read an article for five minutes, you’re still interested in reading that article for five minutes,” Reid noted. The goal is to point users to the right page more accurately, reducing the “bounce” rate where users return to Search because the first result didn’t satisfy their intent.
The Synergy Between AI and the Web
There is a persistent narrative that AI and the web are in competition—a zero-sum game where one must win at the expense of the other. Reid argues that this is a myth. In reality, Google’s data suggests that people want both. AI serves as the starting point, a way to orient oneself in a sea of information, while the web provides the depth and human experience that models cannot replicate.
Human perspective remains a high-value commodity. Whether it is a unique take on a political event, a personal review of a gaming laptop, or a nuanced tutorial on coding, users value the expertise of other people. AI can summarize the general consensus, but it cannot replace the authority of a trusted voice. Reid believes AI helps users “get started” and then makes it easier for them to “dig in” to the actual sources.
The Death of “Keywordese” and the Rise of Natural Language
For decades, users have been trained to speak “computer.” We learned to strip away grammar and context, typing fragmented phrases like “best pizza NYC” or “iPhone 15 specs” into the search bar. This behavior, which some call “keywordese,” was a byproduct of the limitations of early search technology. Users knew that if they asked a complex question, the machine might get confused.
With the integration of Large Language Models (LLMs) into Search, this is changing rapidly. Reid observes that queries are becoming meaningfully longer and more conversational. Users are no longer translating their needs into keywords; they are expressing their full problems. Instead of searching for “clogged drain fix,” a user might now type: “My kitchen sink is draining slowly and there is a metallic smell, what should I do first and what tools do I need?”
Why Longer Queries Benefit the Ecosystem
This shift toward natural language is a significant development for the search ecosystem. When a user provides more context, Google can provide a more accurate and helpful response. This doesn’t just benefit the user; it helps publishers as well. Long-tail queries allow Google to match users with highly specific content that matches their exact intent. This leads to higher quality traffic for websites—visitors who are more likely to stay on the page because the content perfectly addresses their complex query.
Reid views this as a fulfillment of Google’s core mission: making the world’s information not just organized, but truly useful. By doing the “translation” work on behalf of the user, AI allows people to ask more questions and find better solutions to real-world problems.
When Does Google Show an AI Overview?
It is a misconception that AI Overviews will eventually cover 100% of searches. Google is being highly selective about when and where these summaries appear. The decision to trigger an AI Overview is query-dependent and based on a variety of quality signals.
Reid explained that Google avoids forcing AI into the search experience if it doesn’t add clear value. “We don’t want to put an AI Overview if we think it’s not going to be high quality,” she stated. As the underlying models become more powerful, Google can cover more cases, but the focus remains on the “best response” for the specific question. If a standard list of links or a featured snippet is the most effective way to serve the user, Google will stick with the traditional format.
This selective approach ensures that AI is used as a tool for enhancement rather than a default replacement. For many categories—especially those involving navigational queries (like “login to Gmail”) or simple commercial searches—the traditional interface remains the most efficient.
The Future of Monetization: Can Ads Coexist with AI?
One of the biggest questions facing Google’s business model is how AI will affect advertising revenue. If AI provides the answer, will people still click on ads? Reid points out a fundamental truth about search: the vast majority of queries are not currently monetized. In fact, ads appear on less than a quarter of all Google searches. Many of the queries currently being handled by AI Overviews are informational in nature and were never high-revenue generators for the platform.
However, when it comes to commercial intent, the dynamics are different. “The answer doesn’t buy the pair of shoes,” Reid remarked. “You actually have to buy the shoes, right? So you still have to go pick a merchant for that.”
Improving Ads through AI Context
Surprisingly, AI might actually improve the performance of search ads. As users move away from keywords and toward detailed, natural language descriptions of their needs, Google gains a better understanding of what the user is looking for. This allows for more precise ad targeting. If a user describes a specific problem they are having with their home office setup, Google can surface an ad for a product that solves that exact issue, rather than a generic ad for “office furniture.”
Furthermore, as AI makes search more useful, people are likely to search more often. Reid suggests that an expansion of queries leads to an expansion of opportunities. As users become more comfortable using Google for complex, multi-step tasks, they will inevitably engage in more commercial queries along the way.
Measuring Success: The “Return Rate” Metric
While traffic and click-through rates are important, Google is looking at a much more fundamental metric to judge the success of its AI integration: frequency of use. Reid noted that a key signal for the team is whether the changes cause people to come back to Google Search more often.
In the age of mobile apps and social media, “unlocking your phone” to go to Google is a deliberate choice. If AI Overviews make Search more helpful and less frustrating, users are more likely to make Google their primary destination for all types of inquiries. The goal isn’t just to increase the number of searches per session, but to increase the number of times a user thinks to use Search in their daily life.
Search vs. Gemini: Different Tools for Different Needs
Google currently offers multiple AI entry points: the traditional Search interface, the new “AI Mode,” and the standalone Gemini assistant. There is often confusion about the overlap between these products, but Reid clarified that they serve distinct user intents.
Informational vs. Creative Tasks
According to Reid, users tend to migrate toward different tools based on their goals. If a user has an informational query—looking for facts, news, or specific data—they are much more likely to use Search or AI Mode. These tools are grounded in the web’s index and are designed to provide accurate, source-backed information.
In contrast, Gemini is frequently used for creative and generative tasks. This includes writing emails, brainstorming ideas, coding, or transforming text. While there is a segment of the population that “co-uses” both platforms, the distinction between “finding information” and “creating something new” remains a clear boundary in user behavior.
The Complexity of AI Mode
Reid also noted that “AI Mode” within Search tends to attract longer, more complex, and conversational queries. This suggests that users are starting to understand which tool is best suited for their specific needs, choosing the more robust AI interface when they have a problem that requires more than a simple list of links.
Addressing “AI Slop” and the Content Quality Crisis
The rise of generative AI has led to an explosion of low-quality content, often referred to as “AI slop.” This consists of mass-produced articles, often incoherent or factually incorrect, designed solely to rank in search engines and capture ad revenue. Some critics fear that Google will be overwhelmed by this wave of synthetic content, making it impossible to find genuine human perspectives.
Reid’s response to this is grounded in Google’s history of fighting spam. “Before AI slop, there was slop,” she pointed out. “There was human-generated slop.” For as long as search engines have existed, people have tried to game the system with low-quality content farms and keyword stuffing.
Ranking as the Ultimate Filter
The solution to AI-generated spam is the same as the solution to human-generated spam: sophisticated ranking systems. Google’s primary job is to distinguish between high-value information and “slop,” regardless of who or what created it. Reid emphasized that the focus is on maintaining a “very low rate” of spam in search results.
While the volume of slop may increase due to the efficiency of AI, Google’s ability to analyze signals—such as authority, expertise, and user engagement—remains its most powerful defense. The company continues to invest heavily in its core ranking algorithms to ensure that “great information” remains at the forefront. For creators, this means that the formula for success hasn’t fundamentally changed: focus on quality, originality, and providing real value to the reader.
Conclusion: A New Chapter for Search
The insights shared by Liz Reid paint a picture of a search engine in transition, but one that remains deeply committed to its original mission. AI is not being used to replace the web, but to make it more accessible. By filtering out “bounce” clicks, understanding natural language, and refining its ability to fight spam, Google is attempting to create a version of Search that is more intuitive and useful than ever before.
For publishers and digital marketers, the message is clear: the way people search is changing, but the demand for high-quality, human-centric content is not going away. As “keywordese” fades and conversational queries take over, the opportunity to connect with users on a deeper level is expanding. The future of search isn’t just about AI or the web—it’s about how they work together to solve human problems.