Here’s what’s on our plate today:

• 🍽️ The Laboratory: Google’s agentic Search quietly becomes the web’s OS.

• 📰 LinkedIn malware, AI Paint, and Google shopping.

• 🧰 Weekend To-Do: Test agentic tools for search, shopping, and workflows.

• 🗳️ Friday Poll: How comfortable are you with agentic AI shopping?

Let’s dive in. No floaties needed…

The Laboratory:

How Google is quietly rebuilding the internet around AI agents

For over two decades, internet users have relied on Google as their one-stop source of information. For them, the list of blue links, neatly sorted by relevance to their search, made life much simpler. And, for Google, it opened the gateway to advertising revenue that would fuel the growth of its tertiary businesses.

In this ecosystem, enterprises have also fared well. They reached millions, sometimes through advertising and at other times through sheer visibility on Google’s search results page. As for Google, it tried to find the right balance between showing results that users wanted and those it wanted them to see, to justify its advertising business.

However, all this changed with the emergence of artificial intelligence.

AI ends the era of blue links

While OpenAI was wooing users with ChatGPT’s ability to hold human-like conversations, Google used the same technology to overhaul Search. The aim was to shift Search from a long list of links to better compete with chatbots that could find and present information.

And, with Google’s January 2026 shift to an agentic-by-default model, search has quietly stopped being a directory and started acting more like an operator. The result: instead of pointing users to the web, Google now intervenes, executing tasks on their behalf using its understanding of their lives, preferences, and context.

Google’s personal intelligence shift

Google announced the rollout of Personal Intelligence within the Gemini ecosystem in January 2026. This feature allows Gemini to reason across Gmail, Google Search, Photos, YouTube, Drive, and Calendar in order to answer questions and perform tasks using personal context.

Google explicitly positions this system as moving beyond answering queries toward helping users get things done. This marks a fundamental change in how search operates for logged-in users.

Animish Sivaramakrishnan, group product manager of Gemini Personalisation, told the Financial Times that Google's goal has been to move Gemini beyond a simple question-and-answer tool and toward something more adaptive to real-life situations. As he put it, “We’ve always wanted to build a personal assistant that’s actually useful in those ‘life happens’ moments, evolving Gemini from a very transactional assistant to one that knows you better and better over time.”

By drawing on a user's emails, photos, and past conversations, Gemini can tailor responses based on personal context rather than offering generic suggestions.

For example, when planning a holiday, the system can infer preferences such as travel style, family needs, or favored hotels from prior trips and use that understanding to create a more relevant itinerary.

Why visibility no longer means traffic

This shift in Google’s strategy is poised to change how users search for information and how businesses market themselves.

For nearly twenty years, digital marketing followed a familiar formula. Pick the right keywords, earn enough backlinks, and climb the rankings. That logic no longer holds.

In 2026, those signals still matter, but they are no longer the main event. What matters more is how clearly your information is structured and how reliably it can be traced back to its source. AI-driven search now answers many questions directly on the results page. Research from ioVista suggests this will lead to roughly a 30% drop in click-through rates for the traditional number one position.

That sounds like bad news, but there is a catch. The users who do click are not casually browsing. They are usually looking for the original source of the data that the AI referenced. In other words, they want the root, not the summary.

Marketing for machines, not humans

This is why brands now have to think in terms of agentic readiness. A modern website cannot just be readable by humans. It needs to be usable by machines.

Product catalogs, pricing, loyalty programs, and live inventory need to be exposed through structured systems that AI agents can query and act on directly.

As MarTech points out, if product data is not structured and accessible, it effectively does not exist to the agents making purchasing decisions on behalf of users.

That shift also changes how success is measured. Traffic alone is no longer the right benchmark. The more meaningful question is how often an AI model relies on your data to complete a task.

This idea is increasingly referred to as the share of model, a measure of whether your brand is part of the decision-making process inside the machine, not just visible to the human at the end.

Content wins by being the source

What is driving this change is not just how AI disseminates information, but also how content is consumed, which, in turn, is shaping how it is produced.

The kind of content that used to dominate search is quietly losing ground. Long blog posts written to slowly circle a keyword, padded with explanations and repetition, are no longer rewarded.

Google’s AI does not read pages the way people do. It breaks them into smaller semantic units, usually a few hundred tokens at a time, and evaluates each block on its own meaning and usefulness. In this system, clarity beats length every time.

What matters now is not how much you say, but how cleanly you say one thing. A page succeeds when each section communicates a clear idea that can stand on its own without relying on surrounding filler.

Industry experts at I Love SEO describe this shift as provenance scoring. The system favors the original source of information rather than the brand that repackages it most smoothly.

If a company publishes primary material such as original research, technical specifications, or first-hand data, the AI treats that content as a foundational reference point. It becomes something the system can rely on directly.

On the other hand, content that simply summarizes what others have already published carries little value in this environment. Even if it reads well to a human, it adds nothing new for the model to anchor to. As a result, its likelihood of being cited or used by an agent drops sharply.

In practical terms, this means content strategy is no longer about sounding authoritative. It is about actually being the source. The closer you are to the origin of a fact, the more visible you become inside the systems that now decide what gets surfaced and what gets ignored.

Commerce without browsing

However, the most disruptive change may not be search or content, but in how buying itself works.

Agent-to-agent commerce shifts the customer from a person browsing a website to an AI system acting on that person's behalf. Instead of comparing options, filling out forms, or checking policies, a user's agent handles the entire process. It looks for the best fit, confirms availability, and completes the transaction without requiring constant human input.

Gartner expects this kind of task-focused automation to spread quickly, predicting that by the end of 2026, roughly 40% of enterprise applications will include specialized agents. At that scale, purchasing becomes a background process rather than something users actively manage.

Netcore describes this as agentic execution at scale. Brands deploy their own AI systems that represent pricing, inventory, and return rules. These systems interact directly with consumer agents, negotiating terms, confirming stock, and validating policies in real time. No storefront is involved in the traditional sense. The interaction happens machine to machine.

McKinsey’s research shows that technology companies are already ahead of the curve. Nearly a quarter are using autonomous agents in software engineering and IT operations, which makes consumer-facing use cases a natural next step.

For everyday users, the result is a world where buying feels almost invisible. The familiar friction points of e-commerce do not disappear when improved; they disappear because they are no longer needed.

Checkout pages, forms, and confirmations are replaced by a brief approval and a silent exchange of credentials and payment between two AI systems.

Google’s end position

As such, for Google, the shift away from blue links marks the completion of a long transition that began with simple search and advertising. The company is no longer just organizing the world’s information.

It is actively mediating how that information is used, acted on, and transacted. The familiar list of blue links was never the end goal, but a stepping stone toward influence at scale.

However, whether its vision will bear fruition will have to be seen. As AI continues to shape and influence established business models, Google’s positioning appears to be strong. However, there are still regulatory antitrust oversight to contend with.

If Google can overcome the challenges, it will have become the layer that understands intent and executes outcomes, thereby securing its position not as a gateway to the internet, but as its operating system.

TL;DR

• Search is becoming an operator, not a directory. Google’s ‘personal intelligence’ shift means Gemini now reaches into Gmail, Drive, Calendar, Photos & more to answer questions and execute tasks, not just show blue links.
  

• Visibility no longer equals traffic. AI answers more queries directly, cutting clicks to websites, so the real game becomes being machine-readable and becoming a trusted source the models rely on (‘share of model’), not just ranking #1.
  

• Content must be atomized & original. Google’s systems favor clean, structured, provenance-rich chunks of primary data (research, specs, first-party numbers) over fluffy SEO blogs that just repackage others’ work.
  

• Commerce is shifting to agent-to-agent. Brand and consumer agents will increasingly negotiate and transact directly in the background, turning Google from a search gateway into the de facto operating system that interprets intent and executes outcomes.

Headlines You Actually Need

• LinkedIn Malware: Hackers are using LinkedIn DMs to deliver booby-trapped archives that install remote-access malware on high-value targets.

• AI Coloring Tools: Microsoft is rolling out AI-powered coloring-book generation in Paint and faster Copilot text features in Notepad.

• Google Agentic Shopping: Google is turning Search into an AI shopping operator, with Gemini agents quietly orchestrating purchases behind the scenes.  

Friday Poll

Weekend To-Do

• Try Gemini as a personal operator

  Install the Gemini app (Android, iOS via web wrapper) and give it a real task: “Plan a 3-day trip based on my recent emails & calendar,” or “Draft replies to this week’s unread messages.”
  

• Benchmark an AI-native search engine

  Use Perplexity to research a real question you’d normally Google (e.g., a work topic or big purchase), and compare: What sources does it surface, and how few clicks does it take?
  

• Audit your site for ‘agentic readiness.’

  If you run a business, plug your site into Ahrefs Site Audit or Semrush Site Audit and see how clean your structured data, product pages, and internal links are — then fix one concrete issue that would make your content easier for AI agents to parse.

Rate This Edition