AI Search & GEO News

Parsnipp pulls together a regular digest of the GEO / AEO industry news that brands should have top of mind as more customers go to an AI first instead of a brand website, social media, or retailer.

You can also subscribe to our newsletter, The Weekly Harvest, on LinkedIn:

Subscribe on LinkedIn (opens in a new tab)

The industry

AI Search News

Search Engine Land

Google AI Mode ads reach nearly 30% of queries: Study

High CPC searches were far likelier to show AI text ads, while paid placements rarely boosted citations or organic rankings. Google’s AI Mode showed text ads on 29.45% of commercial queries in a new SE Ranking study, less than a year after ads began appearing in AI-generated search answers.

Parsnipp commentary

The key finding for us: paying for AI Mode ad visibility barely overlaps with earning citations or organic rankings — only ~11% of advertiser domains got cited, and ~2% ranked organically for the same keywords. Paid and earned AI visibility are separate games that both need dedicated strategy, not one budget covering both.

Tech Times

Google Fully Replaced Search With AI: Traffic Reduced From Traditional Search

Three days ago, on July 10, 2026, Google completed the most sweeping transformation of its flagship product in the company’s 27-year history. Every Google search query now returns an AI-generated answer as its primary output — written in real time by Gemini 3.5 Flash, the company’s newest inference model.

Parsnipp commentary

AI answers as the default for every query, with a 58% CTR drop for top-ranking pages and some publishers losing 80–97% of traffic — this is the starkest data point yet for why "rank #1" no longer means what it used to. Being cited inside the answer is quickly becoming more valuable than being the blue link underneath it.

Wordpress VIP

Future of the Web 2026: Chapter 1

Brands have been chasing AI visibility for two years. You’ve spent time and budget on it, yet your audience can’t name a single company they think is doing it well. The brands building for the next phase treat their website as the place where AI gets clean data and humans get something worth their time....

Parsnipp commentary

The gap that jumps out: enterprise teams are spending 16+ hours a week on AI visibility, yet 61% of consumers can't name a single brand doing AI well. Investment without differentiation is exactly the trap GEO work needs to avoid — presence isn't the same as being memorable.

Forbes

The Click Is No Longer The Strategy: How Brands Win In A Zero-Click World

For more than two decades, digital marketing was built around one objective: get the click. Search ads pushed users to landing pages. SEO turned visibility into traffic. Social posts pointed people toward blogs, forms, product pages and downloads. The click became the visible proof that attention had turned into intent. But that logic is breaking....

Parsnipp commentary

Brands need to stop measuring success by click-throughs and start measuring whether they're the answer, not just a link.

MediaPost

ChatGPT Keeps Rolling Out Major Ad, Targeting Updates

Advertising features and updates built on artificial intelligence enable advertisers to enhance targeting, bidding, and ad generation for improved return on investments (ROI), but they carry risks of campaign disruption and misattributed citations.

Parsnipp commentary

Custom audience uploads and automated ad variations are pushing ChatGPT ads closer to Google/Meta-style targeting — useful, but experts are already flagging match-rate and attribution glitches (like a client's own link getting swapped for a competitor's). Move fast on these tools, but budget time for monitoring, not just setup.

The Agile Brand Guide

Agile Brand Guide roundup

Yesterday produced a small set of marketing technology announcements, and they describe one subject from four angles: who controls the layer where AI systems decide what a brand’s customers see.  Particular Audience, WSI, 5W, and Vmake Labs each addressed a different surface where that control is moving away from marketers and toward AI platforms.

Parsnipp commentary

The standout item: enterprise AI purchasing decisions are reportedly forming on GitHub and X roughly six months before an RFP even exists. That's a GEO insight in disguise — developer and community platforms are becoming top-of-funnel for B2B buying decisions, not just consumer ones.

TechCrunch

It’s not about Anthropic vs. OpenAI anymore

The U.S. government is set to take an awful lot of control over which AI models get released. Two weeks after the U.S. government pulled Anthropic’s Fable and Mythos models, OpenAI’s new model seems to be headed for the same limbo.

Parsnipp commentary

Regulatory approval delays are now a bigger constraint on frontier model releases than competition between labs. Not directly a GEO story, but it matters for planning: if new model rollouts (and the search/citation behavior changes that come with them) slow down, visibility strategies built around "the latest model" get more stable, not less.

TechCrunch

Anthropic launches Claude Sonnet 5 as a cheaper way to run agents

As shipping agentic capabilities becomes table stakes among foundation model companies, Anthropic is releasing Claude Sonnet 5, a more powerful and agentic version of the lab’s midsize model.

Parsnipp commentary

A cheaper, strong agentic model becoming the free/Pro default matters for us specifically — more people running agents cheaply means more AI-driven research and purchasing happening by default, not as a novelty. Every model that gets cheaper and more agentic widens the audience brands need to be visible to.

B2The7

Zero-Click Search Just Hit 68% — And the Old SEO Playbook Doesn't Cover It

New research from SparkToro and Similarweb dropped this week with a number that should change how you think about search traffic. In the first four months of 2026, 68.01% of U.S. Google searches ended without a single click anywhere.

Parsnipp commentary

The stat worth internalizing: the share of AI Overview citations going to top-10 Google rankers dropped from 76% to 38% in under a year. Rankings and AI citations are decoupling fast — which is genuinely good news for smaller or newer brands that get their content structured and consistent, even without a page-one ranking.

McKinsey & Company

From campaigns to continuous growth: AI capabilities shaping marketing

Capabilities built around insights, creativity, personalization, agentic commerce, and orchestration form the five core pillars of the future of marketing.

Parsnipp commentary

McKinsey's framing matches what we tell clients constantly: 86% of marketers are excited about AI, but fewer than 10% have actually scaled it — the gap is execution, not technology. "Agentic commerce" (optimizing how AI interprets and recommends your products) is explicitly one of their five core capabilities now, not a side experiment.

The Drum

Mark Ritson: Claude is my Cannes bartender. My AI agent is favoring established brands

Mark Ritson revisits the old prediction that perfect information would kill brands and finds the same mistake being made in the age of agentic AI.

Parsnipp commentary

Ritson's point cuts against a lot of GEO anxiety: AI agents show a strong popularity bias, favoring familiar, well-known brands the same way lazy humans do — meaning strong brands get amplified, not flattened.

Marketing Dive

What defines ChatGPT ads? ‘Super intentional’ users, OpenAI tells Cannes

Executives provided a peek behind the curtain of the four-month-old offering, which is leading a wave of discussion around AI at the festival this year.

Parsnipp commentary

OpenAI says ~20% of ChatGPT queries carry direct commercial intent, and 2,000+ brands are already advertising there through partners like Criteo. This is no longer an experimental channel — it has real scale and real measurement partners behind it already.

In the press

Parsnipp News