Sam Altman has graduated from using Google.
“I realized, over the last couple of months, that I don’t use a lot of the services that I used to use. Like I don’t do Google searches anymore, unless…it’s a very specific thing,” he said during a panel at AI Summit in Paris.
He’s not alone—many of us have moved away from search engines, opting to find information via social media or, of course, LLMs like ChatGPT. According to Applause’s 2024 Generative AI Survey, 91 percent of respondents use chatbots for researching, and 81 percent prefer them over search engines for general inquiries. With their ability to understand natural language and synthesize search results, some are saying agents could replace Google within four years.
All of this will have major implications for businesses that currently rely on SEO to drive traffic to their websites, and leaders would be wise to begin preparing now. Here’s how to start thinking about it.
How AI Agents Are Upending Traditional Searches
Say you want to buy a new couch. In the past, you’d probably peruse the web, looking at photos and reviews until you had A) had a pretty good idea of what you wanted or B) became so overwhelmed you just picked one at random. It was a multi-step process, involving a constellation of different retailers and several mind-numbing hours spent comparing features. By the end of it, settling for an overturned cardboard box would begin to sound pretty good.
Now, AI agents have already made this once tedious task easier. By simply describing the couch you want, down to the fabric, measurements, and price, agents can pull together a list of suggestions that meet your criteria, and even complete the transaction—all without skipping around to multiple sites. Instead of you hunting for answers, the agent becomes a personal shopper, researcher, and concierge in one.
Whereas customers tend to search a relatively narrow set of options among brands they’re already familiar with, agents have no limit to the amount of information they can analyze. That democratizes the shopping process, but it also changes the way marketers must think about their approach—in this new world, brands will be appealing to agents, not consumers.
What It Means For Businesses
AI agents are changing the world fast, and there’s no sign progress will slow down. As Altman himself said, “I cannot overstate how much progress we’re going to make in the next 2 years. We know how to improve these models so, so much.”
All businesses, but particularly ecommerce and SaaS businesses, will have to radically adapt in order to remain competitive. One thing they’ll certainly need to do is ensure products are readily findable by AI agents—if they’re not, they’re likely to get passed over. If competitors have successfully made the jump first, they’re the ones who will ultimately snag the customer.
“The reality of AI agents gathering information and bringing it back [to people], that is search in 2025 and beyond. Making sure you are there from an SEO perspective is absolutely vital,” Kelly Metz, chief investment officer at OMD USA, noted during this year’s Consumer Electronics Show, according to Digiday.
While this new form of SEO doesn’t yet have an established name—some refer to it as agentic AI optimization—the important thing is that businesses consider how to position their products in accordance with its principles.
Keeping Up
AI agents don’t have emotional attachments to certain brands—instead, they’ll evaluate products and services based on reviews, quality and features.
Writing for Harvard Business Review, authors Jur Gaarlandt, Wesley Korver, Nathan Furr and Andrew Shipilov point out that companies will have to ensure that their unique strengths, whether that’s product quality, innovation or customer service, are measurable and recognizable by agentic AI. They’ll also need to optimize for the sources agents rely on, like reviews.
“Whereas in the past the number of reviews on Amazon was a major driver of purchase, it will be ever easier for AI agents to synthesize and act based on the aggregated content of these reviews,” they write.
Leaders will need to rethink how they structure their platforms to stay visible in an agent-first world. Product tags and descriptions should be tailored to meet agents’ prompts, and descriptions should also take care to mention how products are used in order to heighten their visibility.
The line to walk will be appealing to both human customers and agents alike. Just as content shouldn’t be stuffed with unnatural-sounding SEO-speak, agentic optimization should be rich enough in structured data for AI agents to parse easily, but still conversational and compelling enough for living, breathing customers. One way to do this is through code—a well-structured API allows agents to autonomously navigate a site, out of view of human users.
Sam Altman is right—the way we search is changing dramatically, and it’s only going to accelerate. In the age of agentic AI, businesses will need to think beyond traditional SEO and embrace a dual-layered approach, optimizing both their content for human readability and their underlying code for machine accessibility. Those who make their sites easy for agents to understand and interact with be the ones that continue to thrive in the AI-driven economy.
Read the full article here
