To search, or not to search, that is the new question.
Have you been tempted to let a chatbot handle your search queries yet? If not, you will be. And it may feel like heaven or perhaps like hell, and most likely, at some point, like both.
Since around 2010, Google has fortified a +/- 90% market share in internet search. Through their ad-driven business model, your attention has been auctioned out to the highest bidder. On October 21st 2025, OpenAI launched their genAI-driven browser, Atlas, to macOS users. This was not the first conversational browser to be unleashed into the market, but with OpenAI's self-reported 800 million weekly active users, the launch of Atlas represents a potentially significant societal shift in how we humans search for information on the internet.
I say 'potentially' because, for OpenAI to reap that potential, the search experience must be attractive to both macOS and Windows users, AND the search results must be consistently perceived to be good enough over time for these 800 million weekly users to replace Google search with ChatGPT Atlas. Atlas is not alone, though. Other examples of genAI-driven browsers include Comet (Perplexity) and Bing (Microsoft), and we can safely assume that more will join the party in the months to come.
Whether you are a fan of Atlas et al or not, Google's monopoly is ripe for disruption. The conventional attention-auction model is about to become prehistoric, and the new battle for your attention is much more subtly hidden in conversational dressing. There is no doubt that AI will play a significant role in the future of search. However, the role of generative AI as a core technology introduces a new set of highly problematic challenges. They may even become your worst cybersecurity nightmare.
Let's start with explaining some of the key differences.
Conventional search engines return a list of results per query. Do you remember those long hours in the library as a student, searching for publications with static keywords? If you didn't get the exact spelling or combination of words right, the search results were as bountiful as an ascetic.
Over time, modern search technologies have significantly improved that experience. Yet, the inclusion of generative AI as core technology represents a fundamental shift in this landscape. Generative AI enables search systems to handle multi-turn and nested queries, produce context-specific outputs across all types of data, and deliver results in a format that mimics human interaction. Doesn't that sound like search heaven?
The SEO rules no longer apply
Conventional SEO (Search Engine Optimisation) was built to rank webpages for keyword queries to win clicks. Conversational search, on the other hand, does not create page rankings, does not use keyword queries, and does not funnel the user toward clicking on search results. Conventional SEO was designed to optimise the probability of winning the attention auction. With conversational search, there is no auction.
The fundamental driver of conversational search is to keep your attention and maintain your engagement for as long as possible, merging ‘find’ with ‘act’ to create a long-term addiction. The underlying business model for these machines is still largely unsolved, which is why your addiction is important to them. Yes, all providers offer various levels of subscription, but the numbers don’t add up. OpenAI alone is heading toward a net loss of about 10 billion USD in 2025. But with 800 million potential addicted users, the rules of engagement change. We are probably going to see attempts at integrating paid positioning in the conversational search engines and browsers. Whether that will succeed remains to be seen.
The most important change, though, is how users receive search results. Where conventional search provided tedious lists of ranked outputs, conversational search provides compelling summaries of the most relevant outputs according to your particular preferences, nicely wrapped in engaging linguistic embellishment. How and where it actually finds the content remains, to a large extent, a well-kept secret. The reliability of the output is about as high as a monkey’s ability to beat the market on financial investments.
However, if you don’t prepare your online presence for genAI-driven search, you are not even on the monkey’s list of potential investments. Frankly, the situation is bonkers, but pretending it will pass without implications for your company will only leave you vulnerable to both lost opportunities and cyberattacks.
Your marketing resources need to understand this and adopt a dual-path strategy. Key elements to consider when optimising online presence for conversational search engines include:
Focus on high-quality content and structured information on the website. Pretty design is irrelevant to genAI-driven search engines
Include credentials, sources, dates, and unique information. Build topical authority.
Make your webpages easy to crawl.
Hopefully, you are now thinking, "isn't crawling our webpages a bad thing?" Yes, it is, if you want to protect your content. The harsh reality is that genAI-driven search engines thrive on this particular dichotomy: do you want to be visible in the conversational search results or not?
The disturbing price tag of conversational search
Conversational search, aka genAI-driven search, comes with three major challenges:
The cybersecurity nightmare is real.
Because of how generative AI and LLMs (Large Language Models) work, the search engine is vulnerable to malign instructions hidden in whatever content the search algorithm processes. These malign instructions can then work their way into the company's systems and documents, causing everything from long-term espionage and ransomware attacks to information theft and reputation damage. For now, the consensus among cybersecurity professionals seems to be to avoid using ChatGPT Atlas (i.e. the browser from OpenAI) on company machines until at least the biggest holes in the Swiss cheese are filled.The sustainability issues are pressing.
GenAI-driven search requires somewhere around 5-10 times more electricity per query than conventional search. This is due both to the technology itself and the fundamental changes in how we search and what we search for. The type of language model (LLM) and its architecture are of great importance to its power consumption. The massive growth in required computing power means massive growth in data centres, which again means substantial pressure on local electricity and water resources. The societal implications are potentially devastating on a human level.Massive mis- and dis- information.
The likelihood of being manipulated is almost as high as the chance of a white Christmas in Norway—never guaranteed, but pretty close. Rhetoric, the art of persuasion, traces back to the philosophers and politicians of ancient Greece and Rome. Building on their insights, humanity has mastered the craft of influence and manipulation. Well, Generative AI is even better, simply because of how it is designed and built. When combining this exceptional capability with the massive amounts of thoughtless nonsense and intentional lies that live out there on the big internet, conversational search engines become mis- and dis- information replicators on steroids.
What does this mean for your business?
Here are three questions for your board or leadership team:
Do you have a company policy on the use of genAI search engines in general and ChatGPT Atlas in particular? If not, you should get your act together as soon as possible.
How prepared are your marketing resources for the disruption of Google's monopoly and the new rules of the online marketing game?
How can you, as a leader, individual, and representative of your company, prepare yourself for the genAI-driven escalation of mis- and dis- information?
The bottom line of search is that we don't know what the bottom line is yet. But one thing seems certain, the verb "to google" will most likely not be part of our grandchildren's vocabulary.
This article is the a part of my series Where Do We Go From Here? AI as the Natural Next Step in Digital Transformation. Want the rest delivered straight to your inbox? Subscribe to my newsletter.
About the Author
Elin Hauge is a business and data strategist, pragmatic futurist, and eternal rebel. With a unique background spanning physics, mathematics, business, and law, she brings fresh and thought-provoking insights into artificial intelligence and other data-driven technologies. Her work connects today’s realities with tomorrow’s possibilities, focusing on demystifying AI, harnessing data power, managing algorithmic risk, and guiding leaders through digital transformation.