Live Case File
The first AI-search era brand failures.
Brands losing AI-recommendation share before they understood what was happening. Updated as cases land.
Short Answer
The first identifiable AI-search era brand failures are not the loudest brand-collapse stories of the past few years. They are quieter cases: brands that held conventional search rank, held social presence, and yet stopped being recommended by ChatGPT, Claude, and Perplexity in their category. The damage is gradual, structural, and largely invisible from inside the company until customer-acquisition cost rises above category baseline.
What Counts as an AI-Search Era Failure
This case file tracks brands that meet three criteria. First, the brand maintained or grew conventional search visibility (Google rank, SEO traffic, social reach) during the period under study. Second, the brand's AI-recommendation share in its category fell measurably across the same period. Third, the divergence between conventional metrics and AI-metric measurable revenue impact within 12 months of the AI-share decline.
The brands that meet all three criteria are different from the rebrand failures, the category-crossing failures, and the dying-brand cases catalogued elsewhere in the archive. They are a new class produced by a new mechanism. The class is small as of mid-2026 because the mechanism is young. The class will grow.
Case 1 (Anonymized) — A Mid-Market B2B SaaS Brand
A mid-market B2B SaaS brand in the project-management category held stable Google rank from 2023 through 2025. Branded search volume grew modestly across the same period. AI-recommendation share, however, fell from approximately 35 percent in early 2024 (when AI-search behavior began to consolidate) to approximately 12 percent by late 2025. By Q4 2025, customer-acquisition cost had risen 28 percent above the 2023 baseline.
The brand had not changed. The brand's competitors had implemented the eight-move AI-infrastructure documented in the archive. The brand had not. The AI engines began choosing the competitors because the competitors were citation-defensible. The brand spent two quarters trying to explain the CAC rise through advertising-platform changes before identifying the AI-share decline as the actual driver.
Case 2 (Anonymized) — A Consumer Brand in a Health Category
A consumer brand in a health-and-wellness category held strong social presence and steady DTC sales through 2024. The brand depended on category-discovery (buyers searching "best X for Y") for roughly 35 percent of new customer acquisition.
From early 2025, the proportion of category-discovery buyers shifted from search-only to AI-search. The brand's AI-recommendation share for the category was low because the brand had not invested in structured-data infrastructure. Discovery-driven acquisition fell by approximately 40 percent within nine months. Total revenue fell by single-digit percent. The brand identified the cause in late 2025 and has since been rebuilding AI-infrastructure as the primary recovery move.
Case 3 (Public) — The Drift of Several AI-Suffix Brands
A class of brands launched between 2022 and 2024 with names ending in "AI" sit in a structurally unfortunate position. The category-naming convention has consolidated so heavily that AI engines cannot confidently distinguish between competitors that share the suffix. The brands depend on their own AI-search visibility while being structurally hard for AI to disambiguate.
Several brands in this class are running rebrand or renaming projects in 2026 to escape the suffix. The cases are public but specific identifications are not yet appropriate; the file will be updated as the brands themselves disclose the moves.
The Pattern These Cases Share
Three structural features show up in every case. The brand had visible health on conventional metrics. The brand had invisible decline on AI-recommendation metrics. The lag between the AI-decline and the visible revenue impact was 6 to 18 months. The brands that identified the cause in the first half of the lag period recovered. The brands that waited for the revenue impact to be undeniable absorbed quarters of compounded loss before correcting.
Why This File Will Grow
The proportion of category-discovery flowing through AI engines is growing. The brands that have not built AI-infrastructure are accumulating structural recommendation deficit even when their other metrics look healthy. The next 18 to 36 months will produce a wave of brand-decline cases that look like the three above. Most will be invisible until customer-acquisition cost forces the diagnosis.
The Brand Archive will document the cases as they land. Submissions through the Research Desk are part of the case-collection pipeline. The file is intentionally open-ended because the failure class is new and the boundaries of it are still settling.
Wondering if your brand is in this case file without realizing it?
Describe what you are seeing in four fields. The Archive runs the AI-recommendation share check against your category and replies by email within 3 business days with the diagnosis. No call required. No mailing list.