Now GPT-5 is with us, search and discovery on the internet is radically transforming and, for the consumer, transforming for the better. For decades traditional search engines, such as Google, have been the front door to the internet and the starting point for online research and shopping journeys.
Consumers used to typing in fragmented keywords, skimming through a list of often useless and sponsored links, reading comparison articles and reviews on random websites, and gradually forming their options. This could feel like a torturous quest, navigating annoying pop-ups as users clicked through various websites until they found something that felt somewhat reliable.
For brands, successfully attracting buyers was largely dictated by one’s position in the search results, achieved through a combination of backlinks and keyword strategy (i.e. traditional search engine optimisation). However, this long-established model is being quietly, but quickly fundamentally dismantled.
Technology is evolving, and consumer behaviour is evolving with it. Rather than typing short phrases into a search bar, one can now engage in full, conversational queries posed to AI tools like ChatGPT, Perplexity, Claude, Grok and Google Gemini. These platforms are capable of understanding context, comparing thousands of options and delivering synthesised, tailored recommendations.
AI Driven Discovery
This evolution is built on a complete redefinition of the discovery process. Rather than presenting a list of vaguely-related results and leaving the final decision-making to the user, AI tools now assume the role of a trusted guide, filtering, interpreting and recommending. All of us can have a personal shopper with taste – our taste.
For consumers, it is a faster, more convenient experience. For brands, it is a new and urgent challenge, if a product is not present in the AI’s answer, it may as well not exist at all.
The rapid adoption of these AI tools – Capgemini reported in January ‘25 that 58% of us have ditched traditional search engines with Gen AI tools for product and service recommendations, a figure growing rapidly, have forced Google itself to change. Today, 60% of Google Searches end without a click.
It’s worth understanding what has changed technically. Unlike traditional search engines, generative AI does not rely on crawling and indexing in the same way. AI draws upon its world knowledge from its training, product feeds, curated sources, and, in some cases, direct partnerships.
With reasoning models, AI engines like Google Overviews or ChatGPT will run tens of parallel queries to your original question simultaneously, some just rephrasing original questions, others adapting it where the AI will try infer the real intention behind questions based off its knowledge of the user, and even some on competitor products/services for comparison. This process will scan hundreds of sources, before providing you with the eventual answer. LLM’s also update their algorithms much more frequently (i.e. the launch of GPT-5 just last week).
New tech needs new techniques
To remain visible, businesses must now engage with an emerging discipline known as Answer Engine Optimisation (AEO) or Generative Engine Optimisation (GEO). While SEO focuses on making content legible to legacy search engine algorithms, GEO is about ensuring that your products and brand are understood and represented accurately within large-language or reasoning models.
Ultimately, executing GEO well, means answering the specific questions that specific customers are asking AI Engines. To start with, tools like Azoma, Profound, or Otterly.AI can help predict how customers are talking to AI Engines, which otherwise are “blackbox systems”.
Once you understand the themes of the questions that your customers are asking AI Engines, brands need to focus on writing content that answers them. This can take a number of forms – blogs, product listing, recipes, or FAQs. In writing this content, the LLM must ensure it is well structured, comprehensive and contextually rich.
Each individual blog or FAQ should be a singular, confident recommendation that could be served up to an AI system responding to a niche user’s question. To optimise best for LLM search, prioritise clarity and intent over keyword density.
Finally, brands will need to focus on showing up in the sources that these AI engines are currently relying on most – Wikipedia and Reddit. Data from analysing millions of citations show that ChatGPT cites Wikipedia in 43% of responses, and Reddit in 27% of responses. Google AI overviews relies on Reddit 24% of the time, and YouTube 27% of the time. Ensuring brands are well represented across these forums is now critical.
Looking over the agentic horizon
The transformation we’re witnessing today is merely the opening act. The next frontier in AI-driven discovery is already taking shape through agentic search and browsers that don’t just answer questions – they act on them.
Perplexity has begun pioneering this evolution with its experimental browser features, moving beyond simple query responses to actually navigating websites, comparing products, and executing tasks on behalf of users. Meanwhile, industry whispers suggest OpenAI is developing a comprehensive browser that would integrate ChatGPT’s reasoning capabilities directly into web navigation, potentially eliminating the traditional distinction between search and action.
These agentic systems represent a fundamental shift from passive information retrieval to active task completion. Rather than simply telling the user which laptop offers the best value, an agentic browser might automatically compare prices across retailers, check availability, read reviews, and even initiate purchase processes – all while explaining its reasoning in real-time.
Convenience versus consideration
For consumers, this promises an unprecedented level of convenience. Imagine asking a browser to “find and book the best Italian restaurant for tonight that accommodates my dietary restrictions,” and watching it seamlessly navigate reservation systems, cross-reference reviews, and complete the booking without requiring a single additional click.
For brands, this evolution presents both an existential threat and an enormous opportunity. Traditional conversion funnels – where customers browse, compare, and gradually move toward purchase – could collapse into single moments of AI-mediated decision-making. If a brand isn’t present in the AI’s consideration set, or worse, if it’s actively filtered out due to poor reputation signals, it will lose customers before they even exist.
The businesses that thrive in this agentic future will be those that optimize not just for discovery, but for AI-mediated action. This means ensuring that a brand’s APIs are accessible, its product data is structured for AI consumption, and that all brand signals consistently communicate value across the digital ecosystem where these intelligent agents operate.
This is not just a trend, its the new normal
It is tempting to view this shift as an emerging trend, another channel to be optimised in parallel with others. It is not. AI-powered Discovery is not a feature or a bolt-on; it is fast becoming the default mechanism through which consumers make decisions.
The new generation of shoppers, especially younger cohorts, are growing up with conversational interfaces as their first point of contact with the internet. For them, the idea of sifting through hundreds of search results is archaic. 93% of Gen Z use at least two AI Engines every week. They expect answers, and soon even the shopping done for them, not just a shopping list of links returned back. The brands that appear in those answers will win not just their attention, but their trust and loyalty.
Brands and retailers must act with urgency. It’s not about simply updating your website or producing new content. A strategic overhaul, restructure of data and integration of AI ecosystems is needed to ensure a brand is represented not merely as an option, but as the most relevant solution.
Discover the world beyond basic search
Quests for answers take too long in the TLDR era. It is no longer about the need to be discovered. What matters in the age of discovery is the mechanism by which discovery takes place.
Traditional search is fading into the background – necessary, but not sufficient. In its place, a new model of decision-making is emerging: intelligent, conversational and context-driven. The brands that adapt will thrive in this new landscape and those that do not may find themselves increasingly absent from the conversation entirely.
