Deep-tech AI company geoSurge has launched from stealth to address a critical new challenge for businesses: the sudden disappearance of major brands from AI-generated search results. The company is pioneering a discipline called corpus engineering to help brands improve and stabilize their visibility within large-language models (LLMs). This launch comes as new research reveals that even dominant brands can vanish overnight between model updates, posing a significant threat in an era of AI-first consumer discovery.
The Shift to AI-First Search
Consumer behavior is rapidly shifting towards AI-driven platforms for information and product discovery, moving away from traditional search engines. Recent studies indicate that over a third of the population in six major countries now uses generative AI weekly, with platforms like ChatGPT processing nearly 2.5 billion queries daily. This fundamental change means that a brand's presence within AI-generated answers is becoming paramount for reaching modern consumers where they begin their journey.
The commercial implications of this trend are substantial, transforming AI into one of the world's most influential recommendation engines. According to Bain & Company, ChatGPT prompt volume surged by nearly 70% in the first half of 2025, with shopping-related queries doubling in the same period. In this new landscape, being surfaced by an LLM is the new form of organic visibility, directly influencing consumer consideration and purchasing decisions.
A New Risk: The Vanishing Brand Phenomenon
Research from geoSurge highlights a startling instability in this new ecosystem, where even household names can disappear between model updates. For instance, the company’s data shows that airline Ryanair is present in GPT-4 flight queries in the UK but is completely absent in GPT-5. Similarly, luxury brands like Chanel and Burberry, competitive in GPT-4, are omitted from GPT-5 results in the US, demonstrating a widespread and unpredictable vulnerability.
Francisco Vigo, co-founder and CEO of geoSurge, explains that these are not isolated incidents but a structural issue. “LLMs don’t pull from a live index,” Vigo stated, clarifying that they generate answers from a compressed internal memory that shifts with every update. This means a brand can transition from high visibility to being completely gone overnight, often without the organization’s awareness, creating a significant and previously unseen business risk.
Corpus Engineering as the Solution
To address this volatility, geoSurge has developed a method it calls corpus engineering, designed to solve the problem at its source. The approach focuses on strengthening how LLMs learn, store, and ultimately surface brand information, thereby creating more stable and reliable generative visibility across different model versions. This proactive technique moves beyond simply monitoring AI behavior to actively influencing it for more consistent outcomes.
Instead of merely reporting on a brand's current standing, geoSurge’s platform shows organizations how they are encoded inside an LLM and identifies areas of weakness. It then provides actionable intelligence on how to reinforce their presence to withstand future updates. To support its launch, the company has also released a flagship research report titled, “When Brands Disappear Overnight: Why Fragile LLM Visibility Is a New Business Risk.”
As consumers increasingly turn to AI for answers, ensuring brand visibility within these models is no longer optional but essential for survival and growth. The launch of geoSurge and its corpus engineering discipline marks a pivotal moment in digital marketing, offering a strategic way to manage this new frontier. For businesses, understanding and optimizing their presence inside LLMs represents the next evolution of securing their place in the consideration set of tomorrow's customer.