Making the internet better for AI agents might make it worse for everyone else

Until now the web has been split between human users and bots, with rules governing the latter’s access. The recent emergence of personal AI agents that can book flights or negotiate prices is creating a third type of visitor. Applications such as OpenAI’s Dots and Meta’s Muse have spurred a rapid rise in traffic generated by agents, tripling between August and September, with Muse reaching five million downloads in its first month.

Many sites—particularly those focused on security—have begun blocking these agents because they do not fit into systems designed to distinguish humans from bots. Amazon, for example, has blocked access to Muse and is suing the startup Perplexity, while platforms like Shopify already offer dedicated checkout lanes for agents.

Experts warn that if demand for agents continues to grow, web architecture may shift to accommodate them more than human users, leading to less user‑friendly interfaces, constant verifications, and tiered pricing based on the intermediary agent.

This transformation matters because it directly affects consumers’ everyday experience and companies’ ability to safeguard data and business models. The regulatory choices made now will determine whether the internet remains a service for people or becomes a highway dominated by automated agents.

Read the original article on MIT Technology Review

This summary is an informational synthesis produced by dataqbs.com. All rights to the original content belong to its author and the cited media outlet. We act solely as curators of technology news and claim no authorship.