The battle for discovery isn't about ranking anymore — it's about being woven into the AI's answer, recommendation, and checkout flow.

The Moment Everything Shifted

On October 31, 2024, OpenAI did something no AI company had done before: it launched a real search product. Not a chatbot that occasionally browsed the web. Not a research preview. ChatGPT Search was a direct, AI-native challenge to Google's quarter-century dominance over how the world finds information. "We're bringing search to ChatGPT," OpenAI announced, embedding real-time web results with source citations directly into conversational responses. For the first time, a non-Google search experience had the scale, user base, and technical infrastructure to meaningfully divert discovery behavior.

The search industry had seen competitors come and go. Bing spent billions. DuckDuckGo built a privacy niche. Neeva tried and folded. But ChatGPT Search was different. It arrived already used by hundreds of millions of people. It didn't ask users to switch search engines — it simply made the AI they were already talking to smarter, more current, and more capable of replacing the Google search box for an expanding set of queries.

What happened next was even more consequential. OpenAI didn't stop at search. It turned ChatGPT into a shopping destination. And in doing so, it triggered a chain reaction that is now forcing Google, Microsoft, Perplexity, publishers, retailers, and every brand with a digital presence to fundamentally rethink what "being found" means.

This is the story of how discovery fragmented — and why the old rules of search optimization no longer apply.

Phase One: The Search Offensive (October 2024 – March 2025)

ChatGPT Search Opens the Front

When ChatGPT Search launched on Halloween 2024, the significance wasn't just the product — it was the precedent. OpenAI became the first AI-native company to position itself as a genuine alternative to Google's core business. The launch included real-time web crawling, source citations, and visual results. For marketers, the implications were immediate: a new surface where brand visibility would be determined not by PageRank but by whether an LLM considered your content authoritative enough to cite.

Google's response came fast. On March 5, 2025, the company announced it was "expanding AI Overviews and introducing AI Mode," built on its Gemini 2.0 upgrade. Reuters reported the same day that Google was testing "an AI-only version of its search engine" — an interface where traditional blue links disappeared entirely, replaced by a conversational AI response. The message was clear: Google wasn't going to let OpenAI define the AI-search experience unchallenged.

The competitive dynamic was now set. Two tech giants, with fundamentally different architectures and business models, were racing to redefine how 5 billion people discover information, products, and services.

Phase Two: The Commerce Invasion (September 2025 – March 2026)

"Buy It in ChatGPT": The Protocol That Changed Everything

If ChatGPT Search opened a new front in the discovery wars, the Agentic Commerce Protocol (ACP) detonated the battlefield entirely.

On September 29, 2025, OpenAI published a landmark announcement: "Buy it in ChatGPT: Instant Checkout and the Agentic Commerce Protocol." This wasn't an incremental update. It was OpenAI's first concrete step toward what it called "agentic commerce" — the ability for users to research, compare, and purchase products entirely inside ChatGPT without ever visiting a retailer's website.

The ACP allowed ChatGPT to display real-time product information, pricing, availability, and — critically — a native checkout experience. Users could go from "What's the best running shoe for flat feet?" to owning a pair of Asics, without a single click to Google Shopping, Amazon, or a brand's own site.

Shopping Research: From Answers to Guidance

Two months later, on November 24, 2025, OpenAI doubled down with shopping research in ChatGPT — research-style shopping guidance that positioned the AI not as a search tool but as a personal shopping advisor. Users could ask complex comparative questions, receive structured recommendations with pros and cons, and follow a conversational thread that narrowed options based on preferences, budget, and constraints.

This was a fundamentally different modality from traditional search. Google showed you ten blue links and let you figure it out. ChatGPT was becoming a decision partner.

Product Discovery: The Full Ecosystem

By March 24, 2026, OpenAI expanded the ACP again with a broader Product Discovery rollout, offering "richer shopping" capabilities. The protocol now supported expanded product catalogs, richer media, merchant integrations, and what OpenAI described as powering "product discovery in ChatGPT" — a phrase that signaled the platform's ambition to own not just the search moment but the entire consideration-to-purchase journey.

In less than six months, OpenAI had built a commerce layer that threatened to disintermediate retailers, comparison shopping engines, and even Amazon's product search dominance.

Google's Counter-Maneuvers: Defending the Fortress

Gemini 3 and the AI-First Search Experience

While OpenAI pushed into commerce, Google accelerated its own AI transformation. On January 27, 2026, the company announced that Gemini 3 now powers AI Overviews, promising what Google called "a seamless new Search experience." The upgrade emphasized follow-up queries, multimodal inputs, and deeper reasoning — essentially trying to make Google's AI Overviews as conversational and helpful as ChatGPT, while retaining Google's unmatched index breadth and real-time data.

Google's AI Mode, first experimented with in March 2025, represented the company's vision of a post-link search interface. As one Google executive told Reuters in December 2025, the company views AI search as an "expansion for web" — not a replacement, but a broader surface that could drive more discovery, not less.

The Regulatory Squeeze

But Google wasn't operating on its own terms alone. On March 18, 2026, Reuters reported that Google was "developing options to allow AI opt-out in search to ease UK concerns" — responding to pressure from the UK's Competition and Markets Authority (CMA). The proposal would give website owners more control over whether their content appears in Google's generative AI features.

This was a significant concession. Google's AI Overviews had faced publisher backlash from the start, with media companies arguing that AI-generated summaries reduced click-through rates and ad revenue. The opt-out proposal acknowledged a new reality: in the AI search era, platforms and publishers would need explicit agreements about how content gets used, cited, and monetized.

The Infrastructure Layer: Who Controls the Crawlers?

Cloudflare Draws a Line (July 2025)

While the platform wars dominated headlines, a quieter but equally consequential development unfolded in the infrastructure layer. On July 1, 2025, Cloudflare introduced tools allowing publishers to control content use for AI training and crawling — essentially giving website owners the ability to manage which AI bots could access their content, and for what purposes.

The significance was enormous. In a world where AI answers depend on training data and real-time crawling, the ability to block or restrict bot access becomes a negotiation lever. Publishers who felt their content was being exploited by AI systems without compensation now had technical tools to push back. Cloudflare's move, combined with Google's opt-out proposal and OpenAI's content partnerships, signaled the emergence of a new content-rights framework for the AI era.

The Crawler Explosion

Cloudflare's own data painted a striking picture. In a July 2025 report, the company documented sharp growth in GPTBot and AI crawler traffic across the web. "From Googlebot to GPTBot," as Cloudflare titled its analysis — the crawler landscape was diversifying rapidly. Google's traditional search crawler was no longer the only bot that mattered. AI crawlers from OpenAI, Anthropic, Perplexity, and others were becoming significant sources of web traffic in their own right.

Wired confirmed this trend in February 2026, reporting that "AI Bots Are Now a Significant Source of Web Traffic" — calling GEO a "new marketing channel for AI-agent discovery." The fragmentation of the crawler ecosystem meant that brands could no longer optimize for a single bot or a single algorithm. Visibility now required being discoverable, crawlable, and citable across a growing constellation of AI systems.

The Publisher Dilemma: Partners or Platform?

Perplexity's Publishers' Program (July 2024)

The question of how publishers participate in — and profit from — AI discovery was partly answered by Perplexity, which on July 30, 2024, launched its Publishers' Program. The initiative tied answer citations directly to publisher partnerships and revenue sharing. When Perplexity cited a publisher's content in an AI answer, that publisher could earn a share of the revenue.

It was a model that acknowledged a hard truth: if AI answers replace clicks, publishers need a new monetization mechanism. Perplexity's approach — essentially creating an attribution economy for AI citations — became a reference point as the industry debated how content creators would survive in a world where users got their answers without visiting source websites.

The Participation Gap

Not all publishers were eager to engage. A Digiday report from August 2025 found that "despite the hype, publishers aren't prioritizing GEO" — many remained unconvinced or under-resourced. The tension was clear: publishers needed AI visibility to maintain relevance, but the path to that visibility was unclear, the ROI unproven, and the risk of further platform dependency uncomfortably familiar.

Reuters reported in November 2025 that "as AI reshapes shopping, US retailers try to change how they're seen online" — adjusting blogs and their Reddit presence specifically for AI visibility. The adaptation was already underway, even if the playbook was still being written.

Microsoft Enters the Fray: The AEO & GEO Playbook

On January 22, 2026, Microsoft made its most significant move yet into the AI search optimization space, publishing "Microsoft's Guide To Winning In AEO & GEO" — a framework that explicitly acknowledged Answer Engine Optimization (AEO) and Generative Engine Optimization (GEO) as distinct disciplines that marketers needed to master.

Microsoft's entry was a watershed moment. As one of the three major search players (via Bing), a major AI investor (via its OpenAI partnership), and the owner of the dominant enterprise software stack, Microsoft's perspective carried unique weight. The guide signaled that GEO wasn't a fringe concept or vendor hype — it was a strategic priority that one of tech's biggest players was actively teaching marketers how to execute.

Microsoft had already been building the measurement infrastructure. On February 10, 2026, the company officially added AI Performance reporting to Bing Webmaster Tools, giving publishers visibility into how often their content was cited in AI responses and which grounding queries triggered those citations. Combined with a November 2025 explainer on "How AI Search Is Changing the Way Conversions are Measured," Microsoft was effectively building the analytics layer for the GEO era.

The New Discovery Landscape: Fragmented, Multi-Modal, and AI-Native

The End of Single-Target Optimization

The cumulative effect of these platform shifts is a discovery ecosystem that looks nothing like the search landscape of 2023. Ahrefs captured this reality in an August 2025 article urging brands toward "Search Everywhere Optimization" — optimizing across multiple discovery surfaces because search has fragmented beyond any single platform's control.

The GEO research corpus tracked by IndexAI shows the same acceleration. Publication volume on GEO topics grew from 3 items in late 2023 to 28 items in Q1 2026 — a nearly 10x increase driven by platform launches, tooling releases, and enterprise adoption. The most consequential coverage clustered around Google's AI search expansions, OpenAI's search and commerce moves, Microsoft's AI performance tooling, Adobe's LLM Optimizer launch, and the emergence of startups like Profound, Otterly AI, Peec AI, and The Prompting Company — all building visibility tools for AI search.

Three Competing Models Emerge

The ecosystem has crystallized around three competing visions of AI discovery:

Google's AI-Augmented Search: Keep the search engine, add AI Overviews and AI Mode, maintain the ad model, and argue that AI expands rather than replaces web discovery. The bet: users still want the breadth of the open web, just packaged more intelligently.

OpenAI's Conversational Commerce: Transform ChatGPT from a chatbot into an end-to-end discovery and purchase platform. The bet: users prefer a single conversational interface that handles research, comparison, and checkout — no website visits required.

Perplexity's Citation Economy: Build a transparent, attribution-first answer engine that shares revenue with publishers. The bet: sustainable AI discovery requires a fair deal with content creators.

Each model demands a different optimization strategy. Each is pulling the industry in a different direction. And each is evolving rapidly enough that any playbook written six months ago is already partially obsolete.

What This Means for the Ecosystem

The Search Monopoly Is Over — Long Live the Duopoly (or Triopoly)

Google's search monopoly hasn't collapsed overnight. Statistically, Google still processes the vast majority of searches globally. But the strategic monopoly — the assumption that Google is the only surface that matters for discovery optimization — is finished.

As Search Engine Land noted in October 2025, the emerging AI Visibility Index was benchmarking brands across prompts in both ChatGPT and AI Mode, treating them as co-equal discovery surfaces. Marketers who optimize only for Google are now leaving ChatGPT visibility — and the commerce transactions that increasingly flow through it — on the table.

Commerce Discovery Is Being Unbundled

The most profound shift is happening in product discovery. Traditional e-commerce search followed a funnel: awareness on Google or social, consideration on brand sites or Amazon, purchase on a retailer's checkout page. ChatGPT's commerce push compresses that funnel into a single conversation.

The implications are seismic for retailers, comparison shopping engines, affiliate marketers, and even Amazon. If a user can research, compare, and buy entirely within ChatGPT, every intermediate surface — Google Shopping, price comparison sites, review blogs, brand websites — faces disintermediation.

Reuters reported in late 2025 that retailers were already adjusting their strategies, modifying content and presence across platforms specifically to be visible to AI systems. The adaptation was reactive, but the direction was clear: commerce discovery was being rearchitected around AI intermediaries.

The Rise of the "Search Everywhere" Strategy

The fragmentation of discovery has spawned a new strategic imperative. Ahrefs' "Search Everywhere Optimization" framework, Salesforce's guidance on teaching AI "who you are," Shopify's merchant guidance for AI crawlers, HubSpot's GEO toolkits — the industry is coalescing around a multi-platform optimization model that treats Google, ChatGPT, Perplexity, Bing AI, and emerging AI surfaces as parallel targets rather than a single hierarchy.

What Brands Should Do Now

The transformation of the discovery ecosystem demands a fundamental shift in how brands approach visibility. Here's what marketing leaders, e-commerce operators, and strategists should prioritize:

1. Audit Your Visibility Across AI Surfaces — Not Just Google

Start measuring where and how your brand appears in ChatGPT, Perplexity, Bing AI, and Google's AI Overviews. Microsoft's AI Performance report in Bing Webmaster Tools and emerging visibility indexes from vendors like Semrush and Adobe's LLM Optimizer can provide baseline data. If you're not visible in ChatGPT's commerce results for your category, that's a strategic gap that needs immediate attention.

2. Optimize for Citation, Not Just Ranking

Traditional SEO optimizes for position on a search results page. GEO optimizes for being cited in an AI-generated answer. That means structuring content for extraction: clear headings, direct answers, evidence-backed claims, and entity-rich markup. As multiple studies in the GEO corpus confirm, the same optimization levers recur: clear structure, direct answers, evidence-backed claims, fresh and crawlable content, and strong entity consistency.

3. Build Third-Party Validation Into Your Content Strategy

AI systems draw on multiple sources to form answers. Google's own documentation emphasizes people-first content, but the AI corpus consistently shows that third-party validation — press coverage, reviews, authoritative references — is a critical input to AI citation. Muck Rack's research frames earned media as a "core input to GEO." Brands need a PR and content strategy that generates the kind of independent credibility signals AI systems weight heavily.

4. Prepare for Conversational Commerce

If your category is one where ChatGPT's Agentic Commerce Protocol is active or likely to expand, you need a direct integration strategy. This means ensuring your product feeds, pricing, availability, and product content are structured for AI consumption. Shopify's guidance on "optimizing your store for AI" — with its focus on product detail pages for both humans and AI crawlers — is a starting template.

5. Implement Crawler and Content Controls

Use Cloudflare's AI crawl control tools or equivalent solutions to actively manage which AI bots can access your content and under what terms. The content you provide to AI systems is increasingly a strategic asset. Treat bot access as a negotiation, not a default.

6. Reallocate Budget Toward Multi-Platform GEO

Digiday reported in March 2026 that marketers are "shifting growing shares of search spending to GEO." The budget reallocation is real and accelerating. Brands should treat GEO as a distinct budget line — not an SEO subcategory — covering multi-platform optimization, AI-specific content structuring, citation measurement, and commerce integration.

7. Build Internal AI-Search Capability

Pfizer's decision to bring SEO and AI search in-house, reported by Digiday in April 2026, represents a broader trend: large brands building internal AI-search capabilities rather than relying entirely on agencies. The expertise required — understanding LLM behavior, citation patterns, prompt engineering for visibility, and AI-commerce integration — is specialized enough that dedicated internal teams are becoming a competitive advantage.

8. Monitor Regulatory Developments

Google's AI opt-out proposal in response to UK CMA pressure is unlikely to be the last regulatory intervention. The EU, UK, and potentially the US will continue scrutinizing how AI systems use publisher content, how commerce flows through AI intermediaries, and what rights content creators retain. Brands need to track these developments because they will shape the rules of visibility in the years ahead.

The Bottom Line

The discovery ecosystem that served digital marketing for two decades — optimize for Google, capture the click, convert on your site — is being fundamentally rearchitected. ChatGPT's commerce push isn't just a new feature. It's a proof of concept for a world where AI intermediaries handle discovery, evaluation, and transaction in a single conversational flow.

Google isn't going away. But it's no longer the only platform that matters. Perplexity isn't the biggest player, but its citation-economy model may prove the most sustainable for publishers. Microsoft isn't the consumer search leader, but its enterprise tooling and AI performance reporting are shaping how the industry measures success.

The winners in this new ecosystem will be the brands that stop thinking about "search optimization" as a single discipline and start treating discovery as a fragmented, multi-modal landscape where visibility in ChatGPT, Google's AI Overviews, Bing AI, and Perplexity each require distinct but coordinated strategies.

The shift from search to shopping inside ChatGPT isn't the end of the story. It's the end of the beginning. The real transformation — agentic commerce at scale, AI intermediaries handling entire purchase journeys, and a discovery ecosystem permanently split across competing platforms — is just getting started.

IndexAI tracks the evolving landscape of Generative Engine Optimization, AI search platforms, and the future of digital discovery. For more analysis, subscribe to our newsletter and follow our research at https://indexai.news.

Developing story. We'll update as new data is validated by the team.