{"id":128695,"date":"2026-05-28T07:03:18","date_gmt":"2026-05-28T07:03:18","guid":{"rendered":"https:\/\/mailinvest.blog\/index.php\/2026\/05\/28\/googles-i-o-2026-shift-search-is-becoming-an-ai-agent-not-a-tool\/"},"modified":"2026-05-28T07:04:52","modified_gmt":"2026-05-28T07:04:52","slug":"googles-i-o-2026-shift-search-is-becoming-an-ai-agent-not-a-tool","status":"publish","type":"post","link":"https:\/\/mailinvest.blog\/index.php\/2026\/05\/28\/googles-i-o-2026-shift-search-is-becoming-an-ai-agent-not-a-tool\/","title":{"rendered":"Google’s I\/O 2026 shift: search is becoming an AI agent, not a tool"},"content":{"rendered":"

<\/a>\n
<\/p>\n

\n

A <\/em>blog post<\/em><\/a> revealed this week by Veronika H\u00f6ller, Head of Demand Technology at Tresorit and co-founder of Your PPC Physician, argues that Google’s bulletins at I\/O 2026 signify a structural break in how the web works – not a product replace. The piece, shared on LinkedIn on Might 25, 2026, three days after publication, shortly drew reactions from practitioners throughout digital advertising and marketing.<\/em><\/p>\n

From reply engine to determination agent<\/h2>\n

H\u00f6ller’s central declare is direct: Google just isn’t updating Search. It’s changing it. In response to H\u00f6ller, the product unveiled at I\/O 2026 just isn’t a extra succesful model of the search engine that has existed for greater than 25 years. It’s one thing qualitatively completely different – a system that “not simply finds info however more and more completes duties, prepares selections, and takes actions on behalf of customers.”<\/p>\n

That framing aligns carefully with what Google itself demonstrated on Might 19, 2026.\u00a0At I\/O 2026, Google introduced persistent background agents<\/a>\u00a0– autonomous programs that function inside Search with out the person submitting a question. An info agent, for instance, can repeatedly monitor residence listings in opposition to a person’s said necessities and ship a notification when a match seems. The person by no means has to return to Search and reformulate a request. That characteristic is ready to launch for Google AI Professional and Extremely subscribers in the summertime of 2026, with no pricing for these tiers included within the Might 19 announcement.<\/p>\n

H\u00f6ller illustrates the behavioral shift with a concrete journey instance. Beforehand, reserving a resort in Barcelona concerned opening 15 tabs, evaluating critiques, checking availability, studying blogs, looking Maps, and watching TikTok critiques. In response to H\u00f6ller, “the agent more and more handles these steps itself.” That compression of a multi-step analysis journey right into a single delegated instruction is, in her framing, the core of what has modified.<\/p>\n

The collapse of the click-based mannequin<\/h2>\n

The implications for the economics of the online are usually not delicate. H\u00f6ller describes a mannequin the place the normal sequence – person searches, Google returns hyperlinks, person clicks, web site convinces, conversion occurs – “grew to become the inspiration of the trendy web.” That precise mannequin, she argues, is now beginning to break.<\/p>\n

Google AI Mode has already surpassed one billion monthly active users globally<\/a>, with queries working on common 3 times longer than typical search queries and complete quantity greater than doubling each quarter since launch. These figures got here from knowledge revealed on Might 19, 2026, alongside the I\/O bulletins. The dimensions issues: a floor utilized by a billion individuals month-to-month, the place the mannequin is being changed and autonomous brokers are being launched, just isn’t an experiment.<\/p>\n

The site visitors knowledge revealed forward of I\/O reinforces H\u00f6ller’s structural level.\u00a0Organic click-through rates for informational queries featuring AI Overviews fell 61% between June 2024 and September 2025<\/a>, from 1.76% to 0.61%, based on Seer Interactive analysis masking 3,119 queries throughout 42 organisations. Paid click-through charges on those self same queries dropped 68%, from 19.7% to six.34%.\u00a0SISTRIX data from March 2026 showed the click-through rate at position one falling from 27% to 11%<\/a>\u00a0when an AI Overview seems – a lack of practically 60% of clicks a top-ranking web page would in any other case obtain.\u00a0Small publishers have already lost 60% of search traffic over two years<\/a>.<\/p>\n

H\u00f6ller’s argument is that the arrival of brokers makes this trajectory irreversible and accelerates it. When AI programs evaluate, filter, prioritize, summarize, e-book, contact, and buy on behalf of customers, she writes, “the position of internet sites modifications routinely.”<\/p>\n

What web sites turn out to be<\/h2>\n

The piece doesn’t declare web sites disappear. What it argues is that their function shifts. In response to H\u00f6ller, web sites more and more turn out to be “belief alerts, knowledge sources, entity hubs, AI verification layers” reasonably than “digital brochures for people.”<\/p>\n

That formulation echoes a priority that has been constructing throughout the trade.\u00a0Google AI agents were already experimenting with table reservations and task completion directly within search results as early as November 2025<\/a>, when John Mueller described agentic options as permitting the AI to “navigate your web site and assist customers with duties there.” In that framing, the web site nonetheless exists – however the customer just isn’t human. The agent navigates on the person’s behalf. Whether or not the web site proprietor sees a session in Analytics is determined by implementation; whether or not they see a conversion could rely upon whether or not their checkout infrastructure integrates with Google’s funds layer.<\/p>\n

Google launched agentic checkout tools for the 2025 holiday season<\/a>, with its product database holding greater than 2 billion listings up to date each hour based on Google’s technical specs. The Common Cart introduced at I\/O 2026 extends that logic throughout Search, YouTube, Gmail, and the Gemini app – a single cross-merchant cart that checks for worth drops within the background as soon as an merchandise is added. The person’s intent has been captured; Google manages the execution.<\/p>\n

Search as a steady working system<\/h2>\n

One of many extra placing components of H\u00f6ller’s evaluation considerations the shift from reactive to steady. Search used to attend. A person formulated a question, submitted it, and obtained outcomes. That discrete, on-demand mannequin outlined what a search engine was.<\/p>\n

Google’s search chief Nick Fox outlined the technical architecture behind this shift in April 2026<\/a>, describing question fan-out, agentic options, and private intelligence in unusually granular element. The background brokers introduced at I\/O 2026 are the consumer-facing product of that structure. They monitor, within the background, for info the person has expressed an curiosity in – with out requiring the person to actively search.<\/p>\n

In response to H\u00f6ller, this makes Search “designed to repeatedly help,” monitoring “preferences, behaviors, contexts, duties, schedules, pursuits.” She describes the strains between Search, Assistant, Browser, Procuring, Productiveness, Commerce, Suggestions, and private working programs as “beginning to collapse.” That framing is in line with\u00a0what Google’s top AI leaders described internally at I\/O 2026<\/a>: Liz Reid, who leads Search, alongside Jeff Dean and Koray Kavukcuoglu described a product trajectory explicitly shifting away from remoted synchronous queries towards asynchronous, background workflows.<\/p>\n

The belief query<\/h2>\n

H\u00f6ller identifies belief because the mechanism AI brokers will use to determine what to floor. That is, she argues, the query that may truly decide industrial outcomes within the coming years – not rankings, not click-through charges, not touchdown web page conversion charges.<\/p>\n

In response to H\u00f6ller, AI brokers will favor programs that seem “constant, reliable, structured, authoritative, machine-readable, contextually dependable throughout platforms.” That shifts the related optimization floor considerably. Beneath that mannequin, Reddit, LinkedIn, YouTube, podcasts, communities, critiques, PR, and creator content material turn out to be essential not due to web optimization worth however as a result of “AI isn’t just studying web sites. AI is making an attempt to grasp status.”<\/p>\n

Research published on PPC Land has documented that AI search traffic from platforms like ChatGPT, Copilot, and Perplexity already converts at substantially higher rates than traditional search traffic<\/a>\u00a0– with sign-up conversion charges of 1.66% from massive language mannequin site visitors in comparison with 0.15% from search. That differential suggests the guests who do arrive through AI referral are pre-qualified in methods conventional clicks are usually not, which is one early knowledge level in favor of H\u00f6ller’s trust-as-ranking-signal thesis.<\/p>\n

Google has been training its deep search agents<\/a>\u00a0utilizing a dual-agent structure – an information generator agent creates question-answer pairs by iteratively looking and gathering info, whereas a separate search agent makes an attempt to resolve the generated questions. The coaching methodology itself displays how the system prioritizes structured, authoritative sources able to supporting multi-step reasoning chains.<\/p>\n

The open net’s stress take a look at<\/h2>\n

H\u00f6ller is direct in regards to the structural stakes. “At present’s web financial system is determined by publishers creating content material, customers clicking, site visitors being monetized,” she writes. “But when AI delivers the reply instantly – and even completes the motion itself – the open net doubtlessly loses its most essential mechanism.”<\/p>\n

She argues the true query just isn’t whether or not web optimization will survive however “how does the web operate when individuals go to fewer and fewer web sites?” That framing has been a recurring concern within the writer neighborhood all through 2024 and 2025. Media leaders surveyed by the Reuters Institute described Google Internet Search site visitors falling from 51% to 27% of referrals between 2023 and the fourth quarter of 2025.\u00a0News publishers have seen Google search traffic drop by half in two years<\/a>, with the steepest declines concentrated in informational verticals like well being and training.<\/p>\n

A PPC Land analysis from April 2026<\/a>\u00a0discovered that 69% of publishers on the Index Trade platform skilled year-over-year declines in advert alternatives all through 2025, with a mean decline of 14% throughout that group. These have been slot requests – the closest proxy for monetisable site visitors – not direct web page views. The mechanism H\u00f6ller describes – AI finishing the motion earlier than the person clicks – is already producing measurable alerts in advert stock knowledge.<\/p>\n

The behavioral dimension<\/h2>\n

H\u00f6ller’s piece is notable for framing the shift as psychological reasonably than purely technical. “Most AI discussions deal with expertise,” she writes. “However the true transformation is behavioral.” Her argument is that after customers internalise that AI can put together selections, save time, filter suggestions, and deal with duties, search conduct itself modifications. Folks will “evaluate much less, analysis much less, click on much less, manually consider much less.”<\/p>\n

That behavioral shift is tougher to measure than CTR knowledge. Nevertheless it issues for anybody constructing digital advertising and marketing infrastructure. If the baseline assumption – that customers wish to consider choices themselves – modifications, the persuasion structure constructed on high of that assumption (touchdown pages, A\/B testing, conversion funnels) could also be optimizing for a person who’s more and more uncommon.<\/p>\n

Google’s own internal sessions at I\/O 2026<\/a>\u00a0included a warning from a Google engineer that AI will multiply software program output by an order of magnitude or extra, and that programs constructed to deal with at present’s volumes is not going to maintain. That warning was directed at developer infrastructure, but it surely applies equally to advertising and marketing measurement programs constructed for human-paced interplay.<\/p>\n

The query H\u00f6ller poses<\/h2>\n

The submit closes with a formulation that has already circulated broadly within the advertising and marketing neighborhood. In response to H\u00f6ller, the longer term query is not “How do I rank on Google?” however “Does AI perceive and belief my enterprise in any respect?”<\/p>\n

Sharon Peredo Sahid, a World Digital Advertising and Product Strategist who commented on the LinkedIn submit, wrote: “I applaud this, assist it and declare myself loyal to this strategy. Very properly described.”<\/p>\n

That response displays how the piece has landed amongst practitioners who’ve been watching the site visitors knowledge and the I\/O bulletins concurrently. The search engine that underpinned digital promoting for greater than twenty years is being rebuilt into one thing that operates on a unique logic completely. The numbers within the site visitors analysis and the structure in Google’s technical periods are in line with H\u00f6ller’s framing. How lengthy the transition takes – and what digital advertising and marketing seems like on the opposite aspect – stays unresolved.<\/p>\n

Timeline<\/h2>\n
    \n
  • June 2024 – September 2025:<\/strong>\u00a0Natural click-through charges for informational queries with AI Overviews fall 61%, from 1.76% to 0.61%, based on Seer Interactive analysis masking 42 organisations<\/li>\n
  • November 2025:<\/strong>\u00a0Google launches agentic checkout instruments;\u00a0product database holds 2 billion listings updated every hour<\/a><\/li>\n
  • November 17, 2025:<\/strong>\u00a0John Mueller broadcasts Search Console modifications and\u00a0confirms Google’s AI agents can navigate websites and complete tasks on behalf of users<\/a><\/li>\n
  • November 2025:<\/strong>\u00a0Seer Interactive publishes research showing organic CTR fell 61% and paid CTR fell 68% on AI Overview queries<\/a><\/li>\n
  • December 2025:<\/strong>\u00a0News publishers have already lost half their Google search traffic over two years<\/a><\/li>\n
  • February 2026:<\/strong>\u00a0Google reveals training methodology for deep search agents using dual-agent architecture<\/a><\/li>\n
  • March 2026:<\/strong>\u00a0SISTRIX knowledge reveals click-through charge at place one falling from 27% to 11% when AI Overviews seem;\u00a0265 million lost organic clicks per month in Germany alone<\/a><\/li>\n
  • April 2026:<\/strong>\u00a0Index Exchange data shows 69% of publishers experienced year-over-year ad opportunity declines throughout 2025<\/a>; small publishers\u00a0lost 60% of search traffic in two years<\/a><\/li>\n
  • April 24, 2026:<\/strong>\u00a0Nick Fox, Google SVP for Search and Maps, describes the technical architecture of AI Mode, query fan-out, and agentic features in published interview<\/a><\/li>\n
  • Might 19, 2026:<\/strong>\u00a0Google broadcasts at I\/O 2026: Gemini 3.5 Flash replaces AI Mode mannequin, persistent background brokers, rebuilt search field, Common Cart, and growth of Private Intelligence to just about 200 nations;\u00a0AI Mode surpasses one billion monthly active users globally<\/a><\/li>\n
  • Might 19, 2026:<\/strong>\u00a0Google I\/O 2026 brings search changes SEOs did not expect<\/a>, together with background brokers that monitor the online with out person prompts<\/li>\n
  • Might 22, 2026:<\/strong>\u00a0Google publishes “Defining the agentic AI period” panel from I\/O 2026;\u00a0Google’s top AI leaders including Liz Reid, Jeff Dean, and Koray Kavukcuoglu describe the technical decisions behind Gemini 3.5 Flash and agentic Search<\/a><\/li>\n
  • Might 25, 2026:<\/strong>\u00a0Veronika H\u00f6ller shares<\/a> her article “Search Is No Longer Search” on LinkedIn; it receives 13 reactions and a pair of feedback inside three days<\/li>\n<\/ul>\n

    Abstract<\/h2>\n

    Who:<\/strong>\u00a0Veronika H\u00f6ller, Head of Demand Technology at Tresorit and co-founder of Your PPC Physician, writing on her private weblog and LinkedIn.<\/p>\n

    What:<\/strong>\u00a0An in depth analytical piece arguing that Google’s bulletins at I\/O 2026 signify a structural transformation of Search from a passive link-retrieval engine into an energetic determination and task-completion agent, with important penalties for the economics of digital advertising and marketing and the open net.<\/p>\n

    When:<\/strong>\u00a0The article was revealed on H\u00f6ller’s weblog and shared on LinkedIn on Might 25, 2026, three days earlier than this report; the I\/O 2026 bulletins it analyses have been made on Might 19, 2026.<\/p>\n

    The place:<\/strong>\u00a0Printed on H\u00f6ller’s private weblog at veronika-hoeller.de and shared through her LinkedIn profile. The occasions analysed happened at Google I\/O 2026, livestreamed from Google’s campus, and are documented throughout a number of PPC Land articles.<\/p>\n

    Why:<\/strong>\u00a0The piece responds to a spot H\u00f6ller identifies between the dimensions of the structural change introduced at I\/O 2026 and the dialog nonetheless taking place within the advertising and marketing trade, which based on her stays centered on AI Overviews, rankings, and CPCs whereas a extra elementary shift – from search engine to digital agent – is already underway.<\/p>\n

    \n


    \n Share this text \n <\/p>\n

    \n
    \n
    \n <\/a> <\/p>\n


    \n
    \n <\/a><\/p>\n


    \n
    \n <\/a><\/p>\n


    \n
    \n <\/a><\/p>\n


    \n The hyperlink has been copied!<\/small> \n <\/p>\n<\/p><\/div>\n<\/p><\/div>\n<\/p><\/div>\n