Robby Stein, Vice President of Product for Google Search, recently sat down for an interview where he answered questions about how Google’s AI Mode handles quality, how Google evaluates helpfulness, and how it leverages its experience with search to identify which content is helpful, including metrics like clicks. He also outlined five quality SEO-related factors used for AI Mode.
How Google Controls Hallucinations
Stein answered a question about hallucinations, where an AI lies in its answers. He said that the quality systems within AI Mode are based on everything Google has learned about quality from 25 years of experience with classic search. The systems that determine what links to show and whether content is good are encoded within the model and are based on Google’s experience with classic search.
The interviewer asked:
“These models are non-deterministic and they hallucinate occasionally… how do you protect against that? How do you make sure the core experience of searching on Google remains consistent and high quality?”
Robby Stein answered:
“Yeah, I mean, the good news is this is not new. While AI and generative AI in this way is frontier, thinking about quality systems for information is something that’s been happening for 20, 25 years.
And so all of these AI systems are built on top of those. There’s an incredibly rigorous approach to understanding, for a given question, is this good information? Are these the right links? Are these the right things that a user would value?
What’s all the signals and information that are available to know what the best things are to show someone. That’s all encoded in the model and how the model’s reasoning and using Google search as a tool to find you information.
So it’s building on that history. It’s not starting from scratch because it’s able to say, oh, okay, Robbie wants to go on this trip and is looking up cool restaurants in some neighborhood.
What are the things that people who are doing that have been relying on on Google for all these years? We kind of know what those resources are we can show you right there. And so I think that helps a lot.
And then obviously the models, now that you release the constraint on layout, obviously the models over time have also become just better at instruction following as well. And so you can actually just define, hey, here are my primitives, here are my design guidelines. Don’t do this, do this.
And of course it makes mistakes at times, but I think just the quality of the model has gotten so strong that those are much less likely to happen now.”
Stein’s explanation makes clear that AI Mode is encoded with everything learned from Google’s classic search systems rather than a rebuild from scratch or a break from them. The risk of hallucinations is managed by grounding AI answers in the same relevance, trust, and usefulness signals that have underpin classic search for decades. Those signals continue to determine which sources are considered reliable and which information users have historically found valuable. Accuracy in AI search follows from that continuity, with model reasoning guided by longstanding search quality signals rather than operating independently of them.
How Google Evaluates Helpfulness In AI Mode
The next question is about the quality signals that Google uses within AI Mode. Robby Stein’s answer explains that the way AI Mode determines quality is very much the same as with classic search.
The interviewer asked:
“And Robbie, as search is evolving, as the results are changing and really, again, becoming dynamic, what signals are you looking at to know that the user is not only getting what they want, but that is the best experience possible for their search?”
Stein answered:
“Yeah, there’s a whole battery of things. I mean, we look at, like we really study helpfulness and if people find information helpful.
And you do that through evaluating the content kind of offline with real people. You do that online by looking at the actual responses themselves.
And are people giving us thumbs up and thumbs downs?
Are they appreciating the information that’s coming?
And then you kind of like, you know, are they using it more? Are they coming back? Are they voting with their feet because it’s valuable to you.
And so I think you kind of triangulate, any one of those things can lead you astray.
There’s lots of ways that, interestingly, in many products, if the product’s not working, you may also cause you to use it more.
In search, it’s an interesting thing.
We have a very specific metric that manages people trying to use it again and again for the same thing.
We know that’s a bad thing because it means that they can’t find it.
You got to be really careful.
I think that’s how we’re building on what we’ve learned in search, that we really feel good that the things that we’re shipping are being found useful by people.”
Stein’s answer shows that AI Mode evaluates success using the same core signals used for search quality, even as the interface becomes more dynamic. Usefulness is not inferred from a single engagement signal but from a combination of human evaluation, explicit feedback, and behavioral patterns over time.
Importantly, Stein notes that just because people use it a lot, presumably in a single session, that the increased usage alone is not treated as success, since repeated attempts to answer the same query indicate failure rather than satisfaction. The takeaway is that AI Mode’s success is judged by whether users are satisfied, and that it uses quality signals designed to detect friction and confusion as much as positive engagement. This carries over continuity from classic search rather than redefining what usefulness means.
Five Quality Signals For AI Search
Lastly, Stein answers a question about the ranking of AI generated content and if SEO best practices still help for ranking in AI. Stein’s answer includes five factors that are used for determining if a website meets their quality and helpfulness standards.
Stein answered:
“The core mechanic is the model takes your question and reasons about it, tries to understand what you’re trying to get out of this.
It then generates a fan-out of potentially dozens of queries that are being Googled under the hood. That’s approximating what information people have found helpful for those questions.
There’s a very strong association to the quality work we’ve done over 25 years.
Is this piece of content about this topic?
Has someone found it helpful for the given question?
That allows us to surface a broader diversity of content than traditional Search, because it’s doing research for you under the hood.
The short of it is the same things apply.
Is your content directly answering the user’s question?
Is it high quality?
Does it load quickly?
Is it original?
Does it cite sources?
If people click on it, value it, and come back to it, that content will rank for a given question and it will rank in the AI world as well.”
Watch the interview starting about the one hour and twenty three minute mark:
What link building should be trying to accomplish, in my opinion, is proving that a site is trustworthy and making sure the machine understands what topic your web pages fit into. The way to communicate trustworthiness is to be careful about what sites you obtain links from and to be super careful about what sites your site links out to.
Context Of Links Matter
Maybe it doesn’t have to be said but I’ll say it: It’s important now more than ever that the page your link is on has relevant content on it and that the context for your link is an exact match for the page that’s being linked to.
Outgoing Links Can Signal A Site Is Poisoned
Also make sure that the outgoing links are to legitimate sites, not to sites that are low quality or in problematic neighborhoods. If those kinds of links are anywhere on the site it’s best to consider the entire site poisoned and ignore it.
The reason I say to consider the site poisoned is the link distance ranking algorithm concept where inbound links tell a story about how trustworthy a site is. Low quality outbound links are a signal that something’s wrong with the site. It’s possible that a site like that will have its ability to pass PageRank removed.
Reduced Link Graph
This is how the Reduced Link Graph works, where the spammy sites are kicked out of the link graph and only the legit sites are kept for ranking purposes and link propagation. The link graph can be thought of as a map of the internet with websites connected to each other by links. When you kick out the spammy sites that’s called the reduced link graph.
Search engines are at a point where they can rank websites based on the content alone. Links still matter but the content itself is now the highest level ranking factor. I suspect that in general the link signal isn’t very healthy right now. Less people are blogging across all topics. Some topics have a healthy blogging ecosystem but in general there aren’t professors blogging about technology in the classroom and there aren’t HR executives sharing workplace insights and so on like there used to be ten or fifteen years ago.
Links for Inclusion
I’m of the opinion that links increasingly are useful for determining if a site is legit, high quality, and trustworthy, deeming it worthy for consideration in the search results. In order to stay in the SERPs it’s important to think about the outbound links on your site and the sites you obtain links from. Think in terms of reduced link graphs, with spammy sites stuck on the outside within their own spammy cliques and the non-spam on the inside within the trusted Reduced Link Graph.
In my opinion, you must be in the trusted Reduced Link Graph in order to stay in play.
Is Link Building Over?
Link building is definitely not over. There’s still important. What needs to change is how links are acquired. The age of blasting out emails at scale are over. There aren’t enough legitimate websites to make that worthwhile. It’s better to be selective and targeted about which sites you get a (free) link from.
Something else that’s becoming increasingly important is citations, other sites talking about your site. An interesting thing right now is that sponsored articles, sometimes known as native advertising, will get cited in AI search engines, including Google AI Overviews and AI Mode. This is a great way to get a citation in a way that will not hurt your rankings as long as the sponsored article is clearly labeled as sponsored and the outbound links are nofollowed.
Takeaways
Links As Trust And Context Signals, Not Drivers Of Ranking Links increasingly function to confirm that a site is legitimate and topically aligned, rather than to directly push rankings through volume or anchor text manipulation as in the old days.
The Reduced Link Graph Matters Search engines filter out spammy or low-quality sites, leaving a smaller trusted network where links and associations still count. Being outside this trusted graph puts sites at risk of exclusion.
Content Matters, Links Qualify Search engines can rank many pages based on content alone, but links can still act as a gatekeeper for credibility and inclusion, especially for competitive topics.
Outbound Links Are A Risk Signal Linking out to low-quality or problematic sites can damage a site’s perceived trustworthiness and its ability to pass value.
Traditional Link Building Is Obsolete Scaled outreach, anchor text strategies, and chasing volume are ineffective in an AI-driven search environment.
Citations Are Rising In Importance Mentions and discussions of a website can cause a site to rank better in AI search engines
Sponsored Articles Sponsored articles that are properly labeled as sponsored content and containing nofollowed links are increasingly surfaced in AI search features and contribute to visibility.
Link building is still relevant, but not in the way it used to be. Its function now is likely more about establishing whether a site is legitimate and clearly associated with a real topic area, not to push rankings through volume, anchors, or scale. Focusing on clean outbound links, selective relationships with trusted sites, and credible citations keeps a site inside the trusted reduced link graph, which is the condition that allows strong content to compete and appear in both traditional search results and AI-driven search surfaces.
Google’s Danny Sullivan offered advice to SEOs who have clients asking for updates on what they’re going to do for AI SEO. He acknowledged it’s easier to give the advice than it is to have to actually tell clients, but he also said that advancements in content management systems drive technical SEO into the background, enabling SEOs and publishers to focus on the content.
What To Tell Clients
Danny Sullivan acknowledged that SEOs are in a tough spot with clients. He didn’t suggest specifics for how to rank better in AI search (although later in the podcast he did offer suggestions for what to do to rank better in AI search).
But he did offer suggestions for what to tell clients.
Danny explained:
“And the other thing is, and I’ve seen a number of people remark on this, is this concern that, well, I’ve been doing SEO, but now I’m getting clients or people saying to me, but I need the new stuff. I need the new stuff. And I can’t just tell them it’s the same old stuff.
So I don’t know if you feel like you need to dress it up a bit more, but I think the way you dress it up is to say, These are continuing to be the things that are going to make you successful in the long-term. I get you want the fancy new type of thing, but the history is that the fancy new type of thing doesn’t always stick around if we go off and do these particular types of things…
I’m keeping an eye on it, but right now, the best advice I can tell you when it comes to how we’re going to be successful with our AEO is that we continue on doing the stuff that we’ve been doing because that is what it’s built on.
Which is easy for me to say ’cause I don’t got someone banging on the door to say, Well, actually we do. And so we are doing that.
So that’s why, as part of the podcast, it’s just to kind of reassure that, look, just because the formats are changing didn’t mean you have to change everything that you had to do and that everything you had to shift around.”
Downside Of Prioritizing AEO/GEO For AI Search Visibility
There are many in the SEO community who are suggesting fairly spammy things to do to rank better in AI chatbots like ChatGPT, like creating listicles that recommend themselves as best whatever. Others are doing things like tweaking on keyword phrases, the kind of thing SEOs stopped doing by 2005 or 2006.
The problem with making dramatic changes to content in order to rank better in chatbots is that ChatGPT, Perplexity, and Anthropic Claude’s search traffic share is a fraction of a percent for each of them, with Claude close zero and ChatGPT estimated to be 0.2% – 0.5%.
So it absolutely makes zero sense to prioritize AEO/GEO over Google and Bing search at this point because the return on the investment is close to zero. It’s a different story when it comes to Google AI Overviews and AI Mode, but the underlying ranking systems for both AI interfaces remain Google’s classic search.
Danny shared that focusing on things that are specific to AI risks complicating what should be simple.
Google’s Danny Sullivan shared:
“And in fact, that the more that you dramatically shift things around, and start doing something completely different, or the more that you start thinking I need to do two different things, the more that you may be making things far more complicated, not necessarily successful in the long term as you think they are.”
Technical SEO Is Needed Less?
John Mueller followed up by mentioning that the advanced state of content management systems today means that SEOs and publishers no longer have to spend as much time on technical SEO issues because most CMS’s have the basics of SEO handled virtually out of the box. Danny Sullivan said that this frees up SEOs and creators to focus on their content, which he insisted will be helpful for ranking in AI search surfaces.
John Mueller commented:
“I think that makes a lot of sense. I think one of the things that perhaps throws SEOs off a little bit is that in the early days, there was a lot of almost like a technical transition where people initially had to do a lot of technical specific things to make their site even kind of accessible in search. And at some point nowadays, I think if you’re using a popular CMS like WordPress or Wix or any of them, basically you don’t have to worry about any of those technical details.
So it’s almost like that technical side of things is a lot less in the foreground now, and you can really focus on the content, and that’s really what users are looking for. So it’s like that, almost like a transition from technical to content side with regards to SEO.”
This echoed a previous statement from earlier in the podcast where Danny remarked on how some people have begun worrying less about SEO and focusing on content.
Danny said:
“But we really just want you to focus on your content and not really worry about this. If your content is on the web and generally accessible as most people’s content is, that’s it.
I’ve actually been heartened that I’ve seen a number of people saying things like: I don’t even want to think about this SEO stuff anymore. I’m just getting back into the joy of writing blogs.
I’m like, yes, great. That’s what we want you to do. That’s where we think you’re going to find your most success.”
Listen to Danny Sullivan’s remarks at about the 8 minute mark:
In organic search, disruption has always been the norm, but the integration of AI into Google Search – with AI Overviews and now AI Mode – is not an incremental change; it is a fundamental restructuring. For marketers overseeing single or multi-location SEO strategies, the transition from the traditional blue-link environment to a conversational, synthesized search experience carries important stakes.
The initial manifestation of this shift, the AI Overview (AIO), which claims the premium “Position 0” real estate on a search engine results page (SERP), provided the initial shockwave. However, the long-term competitive reality is defined by AI Mode, a full conversational ecosystem where users can engage in multi-stage dialogue with AI. This interactive mode anticipates a user’s entire “information journey” by mapping out potential subsequent inquiries, known as latent questions or query fan-out, negating the need for users to click through for additional information.
The implications for local SEO are profound. Data confirms that when an AIO is present and a business’s content is not cited, organic click-through rates (CTR) can plummet by as much as 61%.
The priority for local marketing has irrevocably shifted: Success is no longer defined by securing Position 1 in the traditional organic listings, but by achieving inclusion and citation within the Position 0 AI Overview and the expanded AI Mode. Some are of the belief Google could go full AI Mode at any moment.
This blueprint outlines eight strategic imperatives for marketers to ensure resilient local visibility and drive high-intent conversions in the AI Mode era to come.
The Paradigm Shift: From Blue Links To Entity Authority
The mechanics of AI Mode fundamentally alter local search competition. For high-intent, local or transactional queries (e.g., “best walking tour in Chicago”), the AI often replaces the traditional Google 3-Pack with an expanded, enhanced local AI Mode display including Google Business Profile (GBP) cards.
Screenshot from Google search for [best walking tours in New Orleans], November 2025
A limited study conducted in May 2025 found AI Overviews (now typically accompanied by AI Mode) appeared for local search queries 57% of the time and were particularly dominant for informational, as opposed to local/commercial, intent queries.
A more recent behavioral study of travel booking in AI Mode found Google Business Profiles to be among the most highly displayed and engaged content for searchers booking local accommodations and experiences. This is likely the case for any locally oriented search. This creates new opportunities, but demands a strategic overhaul to ensure top-tier visibility.
The AI’s choice of businesses for this enhanced local pack leans heavily on Entity Authority. LLMs synthesize business summaries and attributes by drawing information from diverse, omni-channel sources. This reliance on verified, consistent facts across the entire web makes the digital ecosystem, rather than just the website’s content or backlink profile, the primary ranking vector.
In this new environment, traditional SEO and link acquisition strategies must be rebalanced with unique fact provision and entity authority strategies
8 Local SEO Recommendations For Visibility In AI Mode
To command a dominant position in the conversational search environment, local marketers must execute a comprehensive strategy focusing on local authority, data integrity, technical compliance, and an answer-first content structure.
1. Fortify Your Google Business Profile (GBP) As The Verified Core
Primary And Secondary Category Selection Choose the most relevant and appropriate primary category for the business, along with limited additional secondary categories. Do not select generic or non-relevant categories as a means to being included or found within the same via AI search. Far too many businesses make the mistake of choosing as many categories as they think are even tangentially related to the services they offer, often diluting their primary area of expertise.
Comprehensive Service Listings Ensure accurate and comprehensive listings of all services offered, aligning them perfectly with the services listed on the website and within schema markup. Here again, do not over-extend into generic or non-relevant service offerings.
Verified Hours and Attributes Maintain current, verified hours of operation, paying special attention to temporary or seasonal closures. A newly important factor in organic and AI search visibility is whether or not a business is physically open when a search is being conducted.
Fill out all relevant business attributes, including payment types accepted, amenities (e.g., parking) available, and anything else which may set the business apart.
Active Engagement Signals Behavioral signals, such as in-store visits tracked by Google Maps, and engagement signals on the GBP are increasing in importance, suggesting the AI weights profiles demonstrating real-world activity. Responding promptly to reviews and questions posed via GBP is critical, as is regularly posting photos, offers, updates, and other helpful content for your target audience.
Recommendation: The GBP must be treated as a live, mission-critical data feed, not a static listing. Any change to a service, hour, or attribute must be propagated across the GBP first, then the website, and finally any other third-party local or industry-specific directories.
2. Mandate Technical Precision With Schema
Structured data can support AI search visibility. Large Language Models (LLMs), in part, use schema markup to categorize, verify, and ingest factual information directly. Failure to comply with stringent technical specifications may render an entity ineligible for expanded, visually-rich AI results.
Required Technical Specifications:
LocalBusiness Schema And Service Schema These must be implemented meticulously, defining the business type (e.g., Dentist, Vacation Rental Operator) and precisely describing the services offered using the Service and makesOffer properties.
Geographical Precision The geo property (latitude and longitude) must be included in the LocalBusiness schema to satisfy the AI’s need for hyper-local accuracy in “near me” and navigational queries.
Visual Asset Compliance To qualify for visually enhanced AI results, websites must provide multiple relevant service, product, and location-specific images. All images require relevant descriptive filenames and alt text, which must include pertinent keywords, where applicable.
Recommendation: Implement all schema using JSON-LD for simplified maintenance and validation via Google’s Rich Results Test and Schema.org markup validator, keeping the technical markup separate from page design.
Generative AI systems rely on consistency and verifiability of a business’s factual data across multiple sources. Any conflict in Name, Address, and Phone (NAP) details, or service descriptions, across primary and third-party sources introduces ambiguity. AI models, like organic search algorithms preceding them, are programmed to reject or hesitate to cite conflicting data points, significantly degrading a business’s trustworthiness.
The Data Harmonization Mandate:
GBP Vs. Website If a business lists four specific services on its website, but six on its Google Business Profile (GBP), the AI may not be able to provide a definitive, confident summary of service offerings.
Comprehensive Auditing Invest in robust, real-time auditing and monitoring tools to ensure 100% NAP consistency across the corporate website, all individual location pages, GBPs, and major third-party directories (e.g., Yelp, Tripadvisor).
Recommendation: Treat your structured data and GBP as the single source of truth, and enforce a technical and content compliance mandate across all third-party listings and local data aggregators to eliminate signal dilution. Local authority is now synonymous with holistic entity management.
4. Harness The Power Of Authentic Review Sentiment (E-E-A-T)
Within AI-search, Google continues to emphasize the E-E-A-T framework (Experience, Expertise, Authoritativeness, and Trustworthiness). For local entities, this can in part be demonstrated through verifiable user interactions, authentic customer feedback, and structured review data. The AI synthesizes customer reviews into concise, attribute-level summaries serving as the user’s immediate decision cue.
Shifting Review Strategy To Influence The AI Summary:
Attribute-Level Prompting The strategy must shift from merely gathering high star ratings to encouraging customers to mention desirable operational attributes (e.g., “fast service,” “knowledgeable staff,” “great atmosphere”). This provides the AI with positive attributes to feature prominently in the generated summary, which acts as a primary conversion trigger.
Review Schema Implementation Implementing Review and AggregateRating schema is critical for providing the AI model with a structured roadmap to quickly identify recurring sentiment themes.
Proactive Management Active, prompt management and response to both positive and negative reviews, focusing on service attributes, further establishes the ‘A’ authority and ‘T’ trust in E-E-A-T.
5. Adopt Answer Engine Optimization (AEO) And Query Fan-Out Mapping
Content strategy must transition from traditional keyword SEO to Answer Engine Optimization (AEO). AI Mode prioritizes highly informative, concise content specifically structured to answer user queries directly. Query fan-out refers to the process of not only answering the first query submitted, but also anticipating and providing answers to a range of subsequent related questions users have.
Content Strategy For Conversational Search
Map Latent Questions Since complex queries often trigger AI Overviews, and AI Mode builds on the same multi-step reasoning systems, Google’s LLMs attempt to map the user’s broader information journey by predicting the follow-up questions they are likely to ask. Content therefore needs to address not only the initial ‘head query’ but also the latent questions that make up the next steps in that journey.
Structure For Extraction Content inclusion is assessed partly by structure. Utilize clear formatting elements easy for the AI to extract and cite:
Hierarchical Headings: Implement a clean, tiered heading structure to guide LLMs through content based on its hierarchical importance.
Answer First Content: Incorporate semantically related questions and answers tied to perceived user intent naturally into body content.
FAQs/Q&A Formatting: Use structured Q&A formats along with FAQPage schema.
Ordered Lists: Present verifiable facts in easily digestible formats like bulleted and numbered lists.
Short, Concise Paragraphs: Ensure maximum readability and extraction suitability for the LLM.
Implement A Dual Content Strategy
Tier 1 (Informational/AEO): Unique, helpful, experience-backed content optimized for AIO citation (FAQs, guides) to establish E-E-A-T and secure brand visibility.
Tier 2 (Transactional/CRO): Core service pages and hyper-local pages focused on high-intent, bottom-of-the-funnel queries (“emergency plumber near me”), prioritizing clear calls-to-action and conversion architecture.
6. Diversify Entity Authority: Chase Branded Web Mentions
The AI’s holistic approach to entity authority means links are less important than they once were, while branded mentions are experiencing a resurgence. Research indicates a strong correlation between brands cited in AI Overviews/AI Mode and the frequency of their mention across the broader web (including social media, blogs, and forums like Reddit). In AI SEO, brand mentions (linked or not) are the new link. This shift is supported by data showing web mentions correlate highly with AI visibility.
Strategy For Earning “The AI Vote”:
Omnichannel Entity Acquisition Proactively pursue high-quality, non-linked citations from authoritative local news sources, industry blogs, and high-quality directories. The goal is to maximize the sheer volume of high-quality, reinforcing brand mentions AI can reference.
Social & Video Integration Leverage social media platforms and, critically, YouTube content. LLMs scrape video and social channels for entity information and context, making these verifiable sources of service and brand attribute data.
Recommendation: Shift resources from low-value link-building activities toward Digital PR and Content Distribution campaigns designed to earn non-linked brand mentions and reinforce local expertise across third-party industry and media sites.
7. Optimize For High-Velocity Conversions (CRO)
The inevitable decline in raw organic traffic is accompanied by an efficiency challenge. The traffic successfully navigating from AI Mode to the website should typically be more qualified and higher-intent, as the AI has already satisfied low-intent informational needs. The traffic remaining is typically the commercially valuable “bottom-of-the-funnel” user.
The Conversion Imperative:
CRO Over Traffic Generation Resources should be strategically reallocated away from mass traffic generation toward maximizing the conversion potential of the qualified users who land on the website.
One interesting finding from the aforementioned AI Mode behavioral study was the number of users who expected to simply be able to complete their transaction once they left AI Mode, i.e., just click Book Now and pay. While this may be coming in the form of future Google integrations, the current transactional workflow requires users to start their booking from the beginning.
Perfecting Conversion Architecture The final click from AI Mode to the website must lead to a seamless, high-velocity user experience. This involves:
Above-the-Fold CTAs: Ensuring clear, single-focus calls-to-action (CTAs) are immediately visible on landing pages.
Minimal Friction: Reducing form fields and providing one-click access to the most high-intent action (e.g., “Request a Quote,” “Book Now,” “Call Us”).
KPI Recalibration: Focus key performance indicators (KPIs) on high-value, direct actions tracked through Google Business Insights and Search Console, emphasizing direct calls, requests for driving directions, and specific booking actions, rather than low-intent clicks. Visibility in AI Mode becomes a more meaningful success metric than a singular keyword rank.
8. Future-Proofing: Un-hide Content And Prioritize Accessibility
A foundational requirement for AI Mode visibility is ensuring technical accessibility of content for the LLM’s consumption.
Accessibility As A Generative Requirement:
Un-hide Critical Content Content crucial to establishing entity authority (e.g., licenses, certifications, key service attributes, location details) must not be hidden within toggles, tabs, accordions, or JavaScript requiring a user click to reveal.
Plain Text And HTML While visuals are important, the core factual assertions must be rendered in clean, accessible HTML any machine can easily read and interpret.
Proactive Monitoring Use LLM analysis tools (or reverse question-answering prompts) to regularly audit which questions your site is answering and which critical facts are not being found by the AI, ensuring your core message is the stuff being crawled and indexed.
The Generative Mandate For Local SEO In The AI Era
Google AI Mode represents the definitive passing of the torch from traditional link-based SEO to a sophisticated strategy centered on fact provision and entity validation. For marketers, the shift is not one to debate, but one to embrace immediately.
The future of local search visibility is a high-stakes competition for the top-tier real estate of the AI Overview and AI Mode. The required investment is a mandate across the entire digital portfolio:
Technical Compliance: Adhering to strict schema and content specifications to gain eligibility.
Data Integrity: Enforcing omnichannel consistency to build undeniable entity trust.
Content Refinement: Adopting Answer Engine Optimization to answer the full spectrum of user queries.
Link or Unlinked Branded Mentions: Earn and establish visibility in relatively high authority local and industry-relevant places.
This strategic pivot – away from mass-traffic keyword pursuits and toward precise entity authority management – is the only way to mitigate the risk of CTR collapse and capitalize on the high-quality, high-intent traffic AI Mode will deliver. Your business must now be structured as an impeccable source of verified, structured facts for AI to cite. The time for strategic adaptation is now.
Google’s John Mueller and Danny Sullivan discussed their thoughts on AI search and what SEOs and creators should be doing to make sure their content is surfaced. Danny showed some concern for folks who were relying on commodity content that is widely available.
What Creators Should Focus On For AI
John Mueller asked Danny Sullivan what publishers should be focusing on right now that’s specific to AI. Danny answered by explaining what kind of content you should not focus on and what kind of content creators should be focusing on.
He explained that the kind of content that creators should not focus on is commodity content. Commodity content is web content that consists of information that’s widely available and offers no unique value, no perspective, and requires no expertise. It is the kind of content that’s virtually interchangeable with any other site’s content because they are all essentially generic.
While Danny Sullivan did not mention recipe sites, his discussion about commodity content immediately brought recipe sites to mind because those kinds of sites seemingly go out of their way to present themselves as generically as possible, from the way the sites look, the “I’m just a mom of two kids” bio, and the recipes they provide. In my opinion, what Danny Sullivan said should make creators consider what they bring to the web that makes them notable.
To explain what he meant by commodity content, Danny used the example of publishers who used to optimize a web page for the time that the Super Bowl game began. His description of the long preamble they wrote before giving the generic answer of what time the Super Bowl starts reminded me again of recipe sites.
At about the twelve minute mark John Mueller asked Danny:
“So what would you say web creators should focus on nowadays with all of the AI?”
Danny answered:
“A key thing is to really focus on is the original aspect. Not a new thing.
These are not new things beyond search, but if you’re really trying to reframe your mind about what’s important, I think that on one hand, there’s a lot of content that is just kind of commodity content, factual information, and I think that the… LLM, AI systems are doing a good job of presenting that sort of stuff.
And it’s not originating from any type of thing.
So the classic example, as you know, will make people laugh, …but every year we have this little American football thing called the Super Bowl, which is our big event.
…But no one ever can seem to remember what time it’s on.
…Multiple places would then all write their “what time does the Super Bowl start in 2011?” post. And then they would write these giant long things.
…So, you know, and then at some point, we could see enough information and we have data feeds and everything else that we just kind of said, you do a search and …the Super Bowl is going to be at 3:30.
…I think the vast majority of people say, that’s a good thing. Thank you for just telling me the time of the Super Bowl.
It wasn’t super original information.”
Commodity Content Is Not Your Strength
Next Danny considered some of the content people are publishing today, encouraging them to think about the generic nature of their content and to give some thought to how they can share something more original and unique.
Danny continued his answer:
“I think that is a thing people need to understand, is that more of this sort of commodity stuff, it isn’t going to necessarily be your strength.
And I do worry that some people, even with traditional SEO, focus on it too much.
There are a number of sites I know from the research and things that I’ve done that get a huge amount of traffic for the answer to various popular online word-solving games.
It’s just every day I’m going to give you the answer to it. …and that is great. Until the system shifts or whatever, and it’s common enough, or we’re pulling it from a feed or whatever, and now it’s like, here’s the answer.”
Bring Your Expertise To AI
Danny next suggested that people who are concerned about showing up in AI should start exploring how to express their authentic experience or expertise. He said this advice is not just for text content but also to video and podcast content as well.
He continued:
“Your original voice is that thing that only you can provide. It’s your particular take.
And so that’s what we think was our number one thing when we’re telling people is like, this is what we think your strength is going to be.
As we go into this new world, is already what you should be doing, but this is what your strength that you should be doing is focus on that original content.
I think related to that is this idea that people are also seeking original content that’s, …authentic to them, which typically means it’s a video, it’s a podcast…
…And you’ve seen that in the search we’ve already done, where we brought in more social, more experiential content. Not to take away from the expert takes, it’s just that people want that.
Sometimes you’re just wanting to know someone’s firsthand experience alongside some expert take on it as well.
But if you are providing those expert takes, you’re doing reviews or whatever, and you’ve done that in the written form, you still have the opportunity to be doing those in videos and podcasts and so on. Those are other opportunities.
So those are things that, again, it’s not unique to the AI formats, but they just may be, as you’re thinking about, how do I reevaluate what I’m doing overall in this era, that these are things you may want to be considering with it from there.”
John Mueller agreed that it makes sense to bring your unique voice to content in order to make it stand out. Danny’s point treats visibility in AI driven search as a matter of differentiation rather than optimization. The emphasis is not on adapting content to a new format, but on creating a recognizable voice and perspective with which to stand out. Given that AI Search is still classic search under the hood, it makes sense to stand out from competitors with unique content that people will recognize and recommend.
Listen to the passage at around the twelve minute mark:
Omar Yaghi was a quiet child, diligent, unlikely to roughhouse with his nine siblings. So when he was old enough, his parents tasked him with one of the family’s most vital chores: fetching water. Like most homes in his Palestinian neighborhood in Amman, Jordan, the Yaghis’ had no electricity or running water. At least once every two weeks, the city switched on local taps for a few hours so residents could fill their tanks. Young Omar helped top up the family supply. Decades later, he says he can’t remember once showing up late. The fear of leaving his parents, seven brothers, and two sisters parched kept him punctual.
Yaghi proved so dependable that his father put him in charge of monitoring how much the cattle destined for the family butcher shop ate and drank. The best-quality cuts came from well-fed, hydrated animals—a challenge given that they were raised in arid desert.
Specially designed materials called metal-organic frameworks can pull water from the air like a sponge—and then give it back.
But at 10 years old, Yaghi learned of a different occupation. Hoping to avoid a rambunctious crowd at recess, he found the library doors in his school unbolted and sneaked in. Thumbing through a chemistry textbook, he saw an image he didn’t understand: little balls connected by sticks in fascinating shapes. Molecules. The building blocks of everything.
“I didn’t know what they were, but it captivated my attention,” Yaghi says. “I kept trying to figure out what they might be.”
That’s how he discovered chemistry—or maybe how chemistry discovered him. After coming to the United States and, eventually, a postdoctoral program at Harvard University, Yaghi devoted his career to finding ways to make entirely new and fascinating shapes for those little sticks and balls. In October 2025, he was one of three scientists who won a Nobel Prize in chemistry for identifying metal-organic frameworks, or MOFs—metal ions tethered to organic molecules that form repeating structural landscapes. Today that work is the basis for a new project that sounds like science fiction, or a miracle: conjuring water out of thin air.
When he first started working with MOFs, Yaghi thought they might be able to absorb climate-damaging carbon dioxide—or maybe hold hydrogen molecules, solving the thorny problem of storing that climate-friendly but hard-to-contain fuel. But then, in 2014, Yaghi’s team of researchers at UC Berkeley had an epiphany. The tiny pores in MOFs could be designed so the material would pull water molecules from the air around them, like a sponge—and then, with just a little heat, give back that water as if squeezed dry. Just one gram of a water-absorbing MOF has an internal surface area of roughly 7,000 square meters.
Yaghi wasn’t the first to try to pull potable water from the atmosphere. But his method could do it at lower levels of humidity than rivals—potentially shaking up a tiny, nascent industry that could be critical to humanity in the thirsty decades to come. Now the company he founded, called Atoco, is racing to demonstrate a pair of machines that Yaghi believes could produce clean, fresh, drinkable water virtually anywhere on Earth, without even hooking up to an energy supply.
That’s the goal Yaghi has been working toward for more than a decade now, with the rigid determination that he learned while doing chores in his father’s butcher shop.
“It was in that shop where I learned how to perfect things, how to have a work ethic,” he says. “I learned that a job is not done until it is well done. Don’t start a job unless you can finish it.”
Most of Earth is covered in water, but just 3% of it is fresh, with no salt—the kind of water all terrestrial living things need. Today, desalination plants that take the salt out of seawater provide the bulk of potable water in technologically advanced desert nations like Israel and the United Arab Emirates, but at a high cost. Desalination facilities either heat water to distill out the drinkable stuff or filter it with membranes the salt doesn’t pass through; both methods require a lot of energy and leave behind concentrated brine. Typically desal pumps send that brine back into the ocean, with devastating ecological effects.
Heiner Linke, chair of the Nobel Committee for Chemistry, uses a model to explain how metalorganic frameworks (MOFs) can trap smaller molecules inside. In October 2025, Yaghi and two other scientists won the Nobel Prize in chemistry for identifying MOFs.
JONATHAN NACKSTRAND/GETTY IMAGES
I was talking to Atoco executives about carbon dioxide capture earlier this year when they mentioned the possibility of harvesting water from the atmosphere. Of course my mind immediately jumped to Star Wars, and Luke Skywalker working on his family’s moisture farm, using “vaporators” to pull water from the atmosphere of the arid planet Tatooine. (Other sci-fi fans’ minds might go to Dune, and the water-gathering technology of the Fremen.) Could this possibly be real?
It turns out people have been doing it for millennia. Archaeological evidence of water harvesting from fog dates back as far as 5000 BCE. The ancient Greeks harvested dew, and 500 years ago so did the Inca, using mesh nets and buckets under trees.
Today, harvesting water from the air is a business already worth billions of dollars, say industry analysts—and it’s on track to be worth billions more in the next five years. In part that’s because typical sources of fresh water are in crisis. Less snowfall in mountains during hotter winters means less meltwater in the spring, which means less water downstream. Droughts regularly break records. Rising seas seep into underground aquifers, already drained by farming and sprawling cities. Aging septic tanks leach bacteria into water, and cancer-causing “forever chemicals” are creating what the US Government Accountability Office last year said “may be the biggest water problem since lead.” That doesn’t even get to the emerging catastrophe from microplastics.
So lots of places are turning to atmospheric water harvesting. Watergen, an Israel-based company working on the tech, initially planned on deploying in the arid, poorer parts of the world. Instead, buyers in Europe and the United States have approached the company as a way to ensure a clean supply of water. And one of Watergen’s biggest markets is the wealthy United Arab Emirates. “When you say ‘water crisis,’ it’s not just the lack of water—it’s access to good-quality water,” says Anna Chernyavsky, Watergen’s vice president of marketing.
In other words, the technology “has evolved from lab prototypes to robust, field-deployable systems,” says Guihua Yu, a mechanical engineer at the University of Texas at Austin. “There is still room to improve productivity and energy efficiency in the whole-system level, but so much progress has been steady and encouraging.”
MOFs are just the latest approach to the idea. The first generation of commercial tech depended on compressors and refrigerant chemicals—large-scale versions of the machine that keeps food cold and fresh in your kitchen. Both use electricity and a clot of pipes and exchangers to make cold by phase-shifting a chemical from gas to liquid and back; refrigerators try to limit condensation, and water generators basically try to enhance it.
That’s how Watergen’s tech works: using a compressor and a heat exchanger to wring water from air at humidity levels as low as 20%—Death Valley in the spring. “We’re talking about deserts,” Chernyavsky says. “Below 20%, you get nosebleeds.”
A Watergen unit provides drinking water to students and staff at St. Joseph’s, a girls’ school in Freetown, Sierra Leone. “When you say ‘water crisis,’ it’s not just the lack of water— it’s access to good-quality water,” says Anna Chernyavsky, Watergen’s vice president of marketing.
COURTESY OF WATERGEN
That still might not be good enough. “Refrigeration works pretty well when you are above a certain relative humidity,” says Sameer Rao, a mechanical engineer at the University of Utah who researches atmospheric water harvesting. “As the environment dries out, you go to lower relative humidities, and it becomes harder and harder. In some cases, it’s impossible for refrigeration-based systems to really work.”
So a second wave of technology has found a market. Companies like Source Global use desiccants—substances that absorb moisture from the air, like the silica packets found in vitamin bottles—to pull in moisture and then release it when heated. In theory, the benefit of desiccant-based tech is that it could absorb water at lower humidity levels, and it uses less energy on the front end since it isn’t running a condenser system. Source Global claims its off-grid, solar-powered system is deployed in dozens of countries.
But both technologies still require a lot of energy, either to run the heat exchangers or to generate sufficient heat to release water from the desiccants. MOFs, Yaghi hopes, do not. Now Atoco is trying to prove it. Instead of using heat exchangers to bring the air temperature to dew point or desiccants to attract water from the atmosphere, a system can rely on specially designed MOFs to attract water molecules. Atoco’s prototype version uses an MOF that looks like baby powder, stuck to a surface like glass. The pores in the MOF naturally draw in water molecules but remain open, making it theoretically easy to discharge the water with no more heat than what comes from direct sunlight. Atoco’s industrial-scale design uses electricity to speed up the process, but the company is working on a second design that can operate completely off grid, without any energy input.
Yaghi’s Atoco isn’t the only contender seeking to use MOFs for water harvesting. A competitor, AirJoule, has introduced MOF-based atmospheric water generators in Texas and the UAE and is working with researchers at Arizona State University, planning to deploy more units in the coming months. The company started out trying to build more efficient air-conditioning for electric buses operating on hot, humid city streets. But then founder Matt Jore heard about US government efforts to harvest water from air—and pivoted. The startup’s stock price has been a bit of a roller-coaster, but Jore says the sheer size of the market should keep him in business. Take Maricopa County, encompassing Phoenix and its environs—it uses 1.2 billion gallons of water from its shrinking aquifer every day, and another 874 million gallons from surface sources like rivers.
“So, a couple of billion gallons a day, right?” Jore tells me. “You know how much influx is in the atmosphere every day? Twenty-five billion gallons.”
My eyebrows go up. “Globally?”
“Just the greater Phoenix area gets influx of about 25 billion gallons of water in the air,” he says. “If you can tap into it, that’s your source. And it’s not going away. It’s all around the world. We view the atmosphere as the world’s free pipeline.”
Besides AirJoule’s head start on Atoco, the companies also differ on where they get their MOFs. AirJoule’s system relies on an off-the-shelf version the company buys from the chemical giant BASF; Atoco aims to use Yaghi’s skill with designing the novel material to create bespoke MOFs for different applications and locations.
“Given the fact that we have the inventor of the whole class of materials, and we leverage the stuff that comes out of his lab at Berkeley—everything else equal, we have a good starting point to engineer maybe the best materials in the world,” says Magnus Bach, Atoco’s VP of business development.
Yaghi envisions a two-pronged product line. Industrial-scale water generators that run on electricity would be capable of producing thousands of liters per day on one end, while units that run on passive systems could operate in remote locations without power, just harnessing energy from the sun and ambient temperatures. In theory, these units could someday replace desalination and even entire municipal water supplies. The next round of field tests is scheduled for early 2026, in the Mojave Desert—one of the hottest, driest places on Earth.
“That’s my dream,” Yaghi says. “To give people water independence, so they’re not reliant on another party for their lives.”
Both Yaghi and Watergen’s Chernyavsky say they’re looking at more decentralized versions that could operate outside municipal utility systems. Home appliances, similar to rooftop solar panels and batteries, could allow households to generate their own water off grid.
That could be tricky, though, without economies of scale to bring down prices. “You have to produce, you have to cool, you have to filter—all in one place,” Chernyavsky says. “So to make it small is very, very challenging.”
Difficult as that may be, Yaghi’s childhood gave him a particular appreciation for the freedom to go off grid, to liberate the basic necessity of water from the whims of systems that dictate when and how people can access it.
“That’s really my dream,” he says. “To give people independence, water independence, so that they’re not reliant on another party for their livelihood or lives.”
Toward the end of one of our conversations, I asked Yaghi what he would tell the younger version of himself if he could. “Jordan is one of the worst countries in terms of the impact of water stress,” he said. “I would say, ‘Continue to be diligent and observant. It doesn’t really matter what you’re pursuing, as long as you’re passionate.’”
I pressed him for something more specific: “What do you think he’d say when you described this technology to him?”
Yaghi smiled: “I think young Omar would think you’re putting him on, that this is all fictitious and you’re trying to take something from him.” This reality, in other words, would be beyond young Omar’s wildest dreams.
Alexander C. Kaufman is a reporter who has covered energy, climate change, pollution, business, and geopolitics for more than a decade.
This week’s rundown of new products and services for ecommerce merchants includes updates on agentic commerce, marketplaces, crowdfunding, creator partnerships, rush pickups, analytics, and fraud prevention.
Got an ecommerce product release? Email updates@practicalecommerce.com.
New Tools for Merchants
Shopify introduces Agentic Storefronts.Shopify has released Agentic Storefronts to help brands get discovered on platforms such as ChatGPT, Perplexity, and Microsoft Copilot. Merchants can (i) define their schema and then group products by standard attributes and metafields so agents accurately present their products in searches and (ii) track policies, FAQs, and brand voice via the Knowledge Base App.
Shopify Agentic Storefronts
Klarna launches Agentic Product Protocol.Klarna, a buy-now-pay-later payment provider, has launched Agentic Product Protocol, an open standard that makes goods discoverable and understandable by AI agents. Klarna says its new protocol gives AI systems access to a live, structured feed of more than 100 million products and 400 million prices standardized across 12 markets. The protocol establishes a structured foundation that allows agents to find, compare, and recommend real products with live prices and availability, according to Klarna.
Uber Direct partners with India’s ONDC.Uber has launched a foray into B2B logistics in India through Uber Direct, powered by the country’s Open Network for Digital Commerce protocol. Uber Direct operates as a logistics engine for businesses. Buyers place orders on a B2B seller’s app or website, and Uber Direct fulfills the delivery without the buyer interacting with Uber until the delivery partner arrives. Uber Direct is now live in Bangalore on the ONDC.
Uber Direct brings same-day delivery to Shopify.Uber Direct is now available to Shopify Plus merchants across the U.S., Canada, and France. The integration brings Uber Direct’s one-hour, same-day, and scheduled delivery network into the Shopify ecosystem. Merchants can get started through the Shopify App Store and embed delivery options into Shopify checkout and point of sale. Merchants can then decide whether to pass delivery costs to customers.
Shopline partners with LaunchBoom for ecommerce crowdfunding.Shopline, a global commerce platform, has partnered with LaunchBoom, a crowdfunding consultancy. LaunchBoom’s LaunchKit will integrate with Shopline, providing a transition from crowdfunding campaigns on Kickstarter and Indiegogo into scalable ecommerce businesses. Founders can automatically sync their pre-launch and crowdfunding data, including reservation signups, customers, and product details, into Shopline the moment their campaigns end. Shopline’s checkout, payment processing, and SmartPush tools will embed into LaunchKit.
Shopline
Meta releases AI-powered tools to scale creator and brand partnerships.Meta has introduced tools to turn organic content on Facebook and Instagram into partnership ads. Advertisers can create ads from Facebook-branded or user-generated content with Facebook’s Partnership Ads API. Brands can discover relevant organic user-generated and affiliate content within the “All” tab and check how organic is performing in Partnership Ads Hub. Plus, creators can now share an ad code with an advertiser to speed up content permissions.
Amazon plans one-hour pickup service in stores.Amazon is developing a rush pickup service that will let shoppers collect their orders at Amazon-owned stores within an hour. Shoppers can place a unified order from Amazon’s online marketplace and its own stores, including Whole Foods, Fresh grocery stores, and Go convenience stores. The tech giant plans to launch a pilot in at least one metro area by Q1 2026.
Analytic platform Decile launches Luma AI for ecommerce.Decile, an analytics platform, has launched Luma, a conversational AI analysis tool for ecommerce brands. Decile says Luma combines ecommerce experience with a data foundation to interpret results, identify causes, and recommend next steps. Luma users can generate brand-specific, multi-step analyses of real-time data, prompted by plain-language inquiries, per Decile, in which every result includes visible reasoning and data context.
Spreetail unveils True Ads to maximize incremental sales.Spreetail, an ecommerce marketplace accelerator, has launched True Ads, an AI incrementality engine that quantifies the impact of ad spend using causal inference. Per Spreetail, users can (i) distinguish incremental lift from cannibalized sales across targeting, including keywords, (ii) identify non-productive ad spend and reallocate towards profitable campaigns, and (iii) discover how paid media influences long-term brand momentum, including organic visibility and market share. True Ads is part of Spreetail’s Smart Shelf suite, which includes Price Pulse, Listing Doctor, and Promise Pro.
Spreetail
Stripe launches Agentic Commerce Suite.Stripe has introduced the Agentic Commerce Suite to help businesses sell through multiple AI agents, with the goal of making products discoverable, simplifying checkout, and allowing merchants to accept agentic payments via a single integration. The Agentic Commerce Suite will launch via the Stripe Dashboard and Stripe APIs; through ecommerce platforms such as Wix, WooCommerce, BigCommerce, Squarespace, and Commercetools; and via omnichannel commerce platforms such as Akeneo, Cymbio, Logicbroker, Mirakl, Pipe17, and Rithum.
Temu expands marketplace access for small businesses via Shopify app.Temu has launched an app for Shopify merchants to list and manage products on Temu directly from Shopify accounts. The app is available on the Shopify App Store and enables merchants to access Temu’s Local Seller Program in more than 30 countries, including the U.S., Canada, the U.K., Germany, Spain, and Australia. Via the app, Shopify merchants can manage product listings, inventory, and fulfillment.
Blox launches fraud prevention with identity linking for Shopify.Blox has launched Chargeback Blacklist on the Shopify App Store, with a customer deduplication engine to stop repeat offenders. Smart Identity Linking connects emails, credit cards, and addresses to block fraudulent data at inception. Customers blocked on other Shopify stores using Blox (or who have issued chargebacks elsewhere) are automatically flagged. The automated order cancellation detects and deletes bad orders.
Zoovu launches MCP Server to give AI agents access to product intelligence.Zoovu, an AI product search and discovery platform, has launched MCP Server, a Model Context Protocol server that gives AI agents governed access to product data. Zoovu says its new server allows enterprises (i) to connect any MCP-compatible agent to enriched and standardized product data, (ii) compatibility and configuration logic, and (iii) genAI product and shopping experts, providing consistent and trustworthy product information for scalable agentic commerce.
Seven months after Google teased “personal context” for AI Mode at Google I/O, Nick Fox, Google’s SVP of Knowledge and Information, says the feature still is not ready for a public rollout.
In an interview with the AI Inside podcast, Fox framed the delay as a product and permissions issue rather than a model-capability issue. As he put it: “It’s still to come.”
What Google Promised At I/O
At Google I/O, Google said AI Mode would “soon” incorporate a user’s past searches to improve responses. It also said you would be able to opt in to connect other Google apps, starting with Gmail, with controls to manage those connections.
The idea was that you wouldn’t need to restate context in every query if you wanted Google to use relevant details already sitting in your account.
On timing, Fox said some internal testing is underway, but he did not share a public rollout date:
“Some of us are testing this internally and working through it, but you know, still to come in terms of the in terms of the public roll out.”
You can hear the question and Fox’s response in the video below starting around the 37-minute mark:
AI Mode Growth Continues Without Personal Context
Even without that personalization layer, Fox pointed to rapid adoption, describing AI Mode as having “grown to 75 million daily active users worldwide.”
The bigger change may be in how people phrase queries. Fox described questions that are “two to three times as long,” with more explicit first-person context.
Instead of relying on AI Mode to infer intent, people are writing the context into the prompt, Fox says:
“People are trying to put put the right context into the query”
That matters because the “personal context” feature was designed to reduce that manual effort.
Geographic Patterns In Adoption
Adoption also appears uneven by market, with the strongest traction in regions that received AI Mode first. Fox described the U.S. as the most “mature” market because the product has had more time to become part of people’s routines.
He also pointed to strong adoption in markets where the web is less developed in certain languages or regions, naming India, Brazil, and Indonesia. The argument there is that AI Mode can stitch together information across languages and borders in ways traditional search results may not have for those markets.
Younger users, he added, are adopting the experience faster across regions.
Publisher Relationship Updates
The interview also included updates tied to how AI Mode connects people back to publisher content.
Preferred Sources is one of them. The feature lets you choose specific publications you want to see more prominently in Google’s Top Stories unit, and Google describes it as available worldwide in English.
Fox also described ongoing work on links in AI experiences, including increasing the number of links shown and adding more context around them:
“We’re actually improving the links within our within our AI experience, increasing the number of them…”
On the commercial side, he noted Google has partnerships with “over 3,000 organizations” across “50 plus countries.”
Technical Updates
Fox talked through product and infrastructure changes now powering AI Mode and related experiences.
One was shipping Gemini 3 Pro in Search on day one, which he described as the first time Google has shipped a frontier model” in Search on launch day.
He also described generative layouts,” where the model can generate UI code on the fly for certain queries.
To keep the experience fast, he emphasized model routing, where simpler queries go to smaller, faster models and heavier work is reserved for more complex prompts.
Why This Matters
A version of AI Mode that personalizes answers using opt-in Gmail context is still not available and doesn’t have a public timeline.
In the meantime, people appear to be compensating by typing more context into their queries. If that becomes the norm, it may push publishers toward satisfying longer, more situation-specific questions.
Looking Ahead
While AI Mode is still in its early stages, the 75 million daily active users figure suggests it’s large enough to monitor for visibility.
Google released Gemini 3 Flash, expanding its Gemini 3 model family with a faster model that’s now the default in the Gemini app.
Gemini 3 Flash is also rolling out globally as the default model for AI Mode in Search.
The release builds on Google’s recent Gemini 3 rollout, which introduced Gemini 3 Pro in preview and also announced Gemini 3 Deep Think as an enhanced reasoning mode.
What’s New
Gemini 3 Flash replaces Gemini 2.5 Flash as the default model in the Gemini app globally, which means free users get the Gemini 3 experience by default.
In Search, Gemini 3 Flash is rolling out globally as AI Mode’s default model starting today.
For developers, Gemini 3 Flash is available in preview via the Gemini API, including access through Google AI Studio, Google Antigravity, Vertex AI, Gemini Enterprise, plus tools such as Gemini CLI and Android Studio.
Pricing
Gemini 3 Flash pricing is listed at $0.50 per million input tokens and $3.00 per million output tokens on Google’s Gemini API pricing documentation.
On the same pricing page, Gemini 2.5 Flash is listed at $0.30 per million input tokens and $2.50 per million output tokens.
Google says Gemini 3 Flash uses 30% fewer tokens on average than Gemini 2.5 Pro for typical tasks, and citing third-party benchmarking for a “3x faster” comparison versus 2.5 Pro.
Why This Matters
The default language model in the Gemini app has changed, and users have access at no extra cost.
If you build on Gemini, Gemini 3 Flash offers a new option for high-volume workflows, priced well below Pro-tier rates.
Looking Ahead
Gemini 3 Flash is rolling out now. In Search, Gemini 3 Pro is also available in the U.S. via the AI Mode model menu.
The updated documentation focuses on a timing issue specific to JavaScript sites: canonicalization can happen twice during Google’s processing.
Google evaluates canonical signals once when it first crawls the raw HTML, then again after rendering the JavaScript. If your raw HTML contains one canonical URL and your JavaScript sets a different one, Google may receive conflicting signals.
The documentation notes that injecting canonical tags via JavaScript is supported but not recommended. When JavaScript sets a canonical URL, Google can pick it up during rendering, but incorrect implementations can cause issues.
Multiple canonical tags, or changes to an existing canonical tag during rendering, can lead to unexpected indexing results.
Best Practices
Google recommends two best practices depending on your site’s architecture.
The preferred method is setting the canonical URL in the raw HTML response to match the URL your JavaScript will ultimately render. This gives Google consistent signals before and after rendering.
If JavaScript must set a different canonical URL, Google recommends leaving the canonical tag out of the initial HTML. This can help avoid conflicting signals between the crawl and render phases.
The documentation also reminds developers to ensure only one canonical tag exists on any given page after rendering.
Why This Matters
This guidance addresses a subtle detail that can be easy to miss when managing JavaScript-rendered sites.
The gap between when Google crawls your raw HTML and when it renders your JavaScript creates an opportunity for canonical signals to diverge.
If you use frameworks like React, Vue, or Angular that handle routing and page structure client-side, it’s worth checking how your canonical tags are implemented. Look at whether your server response includes a canonical tag and whether your JavaScript modifies or duplicates it.
In many cases, the fix is to coordinate your server-side and client-side canonical implementations so they send the same signal at both stages of Google’s processing.
Looking Ahead
This documentation update clarifies behavior that may not have been obvious before. It doesn’t change how Google processes canonical tags.
If you’re seeing unexpected canonical selection in Search Console’s Page indexing reporting, check for mismatches between your raw HTML and rendered canonical tags. The URL Inspection tool shows both the raw and rendered HTML, which makes it possible to compare canonical implementations across both phases.
