Google I/O 2026: What you need to know for SEO /AEO
Google did not announce that AI mode is becoming the default search experience, but what they did announce does not bode well for those who rely on search clicks.
This week’s newsletter is sponsored by North Star Inbound and Airops
Paid subscribers can see all my videos and photos from I/O at the end of this email
As I predicted recently, AI mode is not becoming the default search experience, but what Google did announce at I/O does not bode well for those who rely on search traffic. Google shared that AI Mode is used by 1 billion people a month and that its queries are doubling every quarter. We can never be certain that AI Mode will always remain a browser click away.
Google will likely not launch AI Mode as a default experience until they are certain it will not impact any monetization efforts, but the product's popularity means they are poised to broaden the push anytime they feel threatened.
[Sponsored by Airops]
2026 AEO Playbook: The 4 plays behind Ramp, Carta, and Webflow’s AI search strategy
Most teams are still measuring AI search the way they measured Google. That’s the part of this shift that keeps tripping people up.
Ranking and being cited aren’t the same job. A page can sit at position one and never get pulled into a single AI answer. Another can rank on page two and get cited in ChatGPT, Gemini, and Perplexity in the same week.
What changed is the selection logic. Answer engines aren’t ranking pages. They’re choosing which source feels safe to quote. That decision runs on structure, evidence, freshness, and how often your brand shows up next to the right concepts elsewhere on the web.
AirOps analyzed 15M+ AI queries to figure out what actually earns citations. A few patterns stood out to me. Pages with clear heading hierarchy are 2.8x more likely to be cited. 85% of brand mentions in AI answers come from off-site sources, not your own domain. And 70% of cited pages were updated in the last 12 months.
The playbook breaks down the 4 structural plays behind how Ramp, Carta, and Webflow are operationalizing this, plus a 90-day action plan to put it in motion. Most useful thing I’ve read on the topic.
I was privileged to attend Google I/O once again to hear live announcements about what Google is launching; however, the most valuable aspect of attending I/O is seeing 1:1 demos of the new products from the product managers who created them and asking direct questions.
Most of you have likely already seen the headlines and read the stories about this being the biggest change to search in many years. I can’t say I agree with that sentiment, because the changes Google launched have already been quietly available to many people for months. Google search is NOT changing. It already changed.
Interestingly enough, the topic of search didn’t come up until about 45 minutes into the morning keynote, which, in my opinion, is itself a signal about the priority of the search product in Google’s ecosystem. When they finally introduced it, they said the search box would now be called “Ask Google” rather than “search.” This is what the media has been breathlessly calling the greatest change in Google search history. This new CTA is mostly a subtle change that doesn’t impact the user experience, since any search you conducted before was already asking Google.
The search bar could handle queries or prompts, and AI mode and Overviews were just one click away. As part of this naming change, Google shared that in AI Overviews, you could directly enter an AI Mode follow-up conversation, which has been the case for everyone in this lab, and most readers were probably in it. Now it is available to everyone.
I don’t think this will have an additional impact on downstream search clicks because anyone who was already doing top-of-funnel discovery-type searches that could have been satisfied with deeper questions to AI was likely already not ready to click off to websites.
Google also announced that, within AI Mode, it was expanding personal intelligence globally. Personal intelligence in AI Mode-only has a benign impact on SEO, but it will have drastic consequences once it reaches general search results.
Wearables
As Google previewed and many have predicted, Google updated its wearable tech with a new glasses partnership with Warby Parker and Gentle Monster. I had the opportunity to demo this product, and while I think it is interesting from a technology standpoint, the core functionality will not disrupt search and will only replace what you can already do on a phone. This is just the hands-free version. I was surprised that Google did not mention Project Astra (see this post) again, which seems like an important part of their glasses strategy.
Stay tuned for a future newsletter that will dive into this demo and experience.
New search changes
In addition, Google shared four new changes to search, which I had a chance to see up close. On their own, you might not think they would impact search clicks, but this is just a start. Except for the last feature below, which currently has no comparison other than YouTube videos, these are new features that allow users to have their needs met directly in search, without visiting a website.
Rather than deciding what to do or where to spend their money after visiting websites, Google will make those decisions for them directly in search results. The ramifications of will could be profound.
Search agents
Users can build their own agents directly in AI Mode, eliminating the need to visit websites and conduct manual research. These agents, which appear to be dynamic and advanced Google alerts, can help users discover when something has changed (like an event being announced) or perform dynamic lookups, such as finding stocks whose prices have changed.
Agentic Booking
Rather than conducting their own research about businesses to frequent, a user can create an agent to find a business that suits their needs. Google will use publicly available information to recommend the business, and if the answers are unknown, it will even call the business to make the requests directly in AI mode.
Agentic coding
Building on the two prior features of using agents to search the web, users can build their own apps directly in search that dynamically plan their days and even coordinate with other Google apps like calendars.
For a user who likes to spend their weekends hiking, this will pull traffic away from sites like Alltrails, and for planning restaurants, this will take traffic away from sites like Yelp.
Generative UI in search
This is the only feature that currently has no parallel, and I found the demos to be fascinating. For answers that require visualizations to truly understand a topic, Google will have experiences to help answer a question. The example query I saw was helping a user understand how their watch winds up. This topic is something you can certainly read about in text, but you get a much deeper understanding when you can visualize and experiment with the result. This feature will be very helpful for students of any age.
What does this mean for SEO/AEO?
AI and not just LLMs, which answer text queries, are here to stay, which means that if you were reliant on clicks from rankings on traditional search results, your numbers will inevitably continue to decline. Many of the features Google previewed are good for users and will help users learn and transact more efficiently than ever.
If your business is service- or transactional-based, users looking to engage with you will still do so even if the funnel has changed. As long as you are the best fit for your customer, and not just market yourself as the best fit, you are well-positioned to continue to have customers directed to you from Google.
There is no better way to force Google to send these users to you than to be the best at what you do.
If your product is informational in nature, you likely have already seen dramatic impacts on traffic, and these changes mean that traffic might continue to decline as Google maintains informational responses in search results. If you are in the fortunate position of having proprietary data and information, you will need to ensure that Google is aware of this information by doing great SEO (technical + PR) with an optimization focus on the follow-up, and that Google's new ways of presenting information still recommend your site for follow-up learning.
Last week, Google released a guide on optimizing for generative features, and I think it was ten years too late. Not only does it recommend thinking of generative AI visibility as an extension of SEO, but it also gives its viewpoint on how to optimize for traditional SEO today.
Provide a unique point of view
Create non-commodity content
Organize content in a way that helps your readers
Add high-quality images and video
Focus on what your users want, and avoid overdoing it.
That last point should guide all SEO and AEO. Rather than fear what is to come, worry about the users. And yes, search funnels will change, but this advice is timeless.
If you have specific questions about any of the features above, I would be happy to hear from you and try to share more about what I learned.
[Sponsored by North Star Inbound ]
What happens when you combine best-in-class SEO with conversion optimization?
BigRentz’s traffic increased by 186% (85k), added 1950 conversions in 12 months.
Self Financial’s traffic increased by 50k/month, and added 685 new customers.
Lastly, Secure Data added 1968 phone calls.
North Star Inbound’s SEO strategies earn leads, conversions, and revenue..
Book a call for a free content audit and 10% off any engagement.
Intrigued? [Learn more here]