Ads that target AI agents rather than humans may sound like something ripped from the pages of science fiction, but it’s a concept that’s gaining traction among marketers thanks to recent thoughts from Perplexity CEO, Aravind Srinivas. On “contrarian marketing” podcasthe painted a picture of a future in which “the user never sees advertising. Unlike Google, different merchants don’t compete for users’ attention. They compete for the attention of AI agents.
Naturally, Srinivas’ comments sparked a torrent of hot takes. Somewhere in the swirl of opinions, a few recurring arguments emerge – both for and against this seemingly far-fetched, but not entirely implausible, view.
Case for advertisements broadcast to AI agents
It provides a cleaner user experience
People are overwhelmed by the endless barrage of ads, especially those that completely miss the mark. But if AI agents become the new target for advertisers, the constant stream of ads could disappear from view altogether. The result? A cleaner, ad-free user experience, allowing consumers to enjoy the web and their platforms without the usual interruptions.
Speaking of user experience, AI agents can create efficiencies for people
By giving AI agents manual or menial tasks, it allows individuals to focus on more sophisticated tasks.
“If you are a consumer, you can, for example, book travel, transport, reservation or buy something,” said Rohith Kolluri, co-founder of software development company Stealth, which is currently building an AI agent for D2C growth marketing teams. . “Wherever a consumer had to click on a link, enter a certain amount of information and complete a series of manual steps, the AI agent could carry out these actions for them. And as this AI gets to know them better, it can interact with the ads itself and make decisions for them.
AI agents can also create efficiencies for B2B operations
Since Kolluri is already in the early stages of creating an AI agent for D2C growth marketing teams, he is very familiar with how it can improve the teams’ performance and capabilities.
“We’re building it to help them with the complete creative analysis, testing and optimization workflow they need to grow their business,” Kolluri said. “It’s basically like adding a junior employee to every team who works 24/7 on this workflow. They have much more leverage and can do many more things.
Case against ads broadcast to AI agents
Advertisers will need to move from targeting humans to targeting AI agents.
For years, marketers have complained about spending money targeting bots instead of real people. Ironically, that may soon be the goal. In the future, Srinivas predicts, rather than enticing masses of consumers to buy their products, advertisers may need to charm AI agents into introducing their products to any user. It’s the ultimate plot twist: brands working to impress the algorithms they once cursed.
“If advertisers can’t do it, or if it doesn’t work, they won’t pay to advertise on these systems because they need to get a return to justify advertising these systems,” Kolluri said. “These companies need to figure out how to create a great user experience, while still allowing advertisers to generate revenue.”
Advertisers will likely need to focus on reaching other platforms.
Since advertisers will target AI agents, they will no longer target mass consumers on these platforms, removing the top of the funnel altogether.
“Platforms with AI agents are going to capture people searching for things they already want,” Kolluri said. “They [advertisers] They will likely need to use different demand generation channels, such as meta ads for example, if they want to increase brand awareness rather than trying to reach these platforms.
Consumers are unlikely to adopt this new behavior anytime soon
What makes sense for the tech world usually requires the rest of the world to catch up. Think about the Metaverse – while Meta CEO Mark Zuckerberg was enthusiastic about this being the future, when adoption didn’t follow as quickly as the company hoped, they have flipped the narrative to suggest this is a longer-term project. project when thinking about adoption.
“I think we’re far from the point where enough people will feel comfortable allowing an AI agent to spend money on their behalf and take control of decision-making,” said a marketing manager for a global bank.
This is a view shared by Tom Limongello, co-founder of Nextipedia. “Unless I train multiple agents and know which ones do what, I’m still going to want to see these ads,” he said. “The machine [AI agent] will not be convinced. It will simply be a matter of trying to break down what exists and characterize it in terms.
