Web traffic from AI agents tripled between August and September as personal assistant programs capable of booking flights and hotel rooms flooded the internet at scale, according to an October 2026 report published in MIT Technology Review's The Algorithm newsletter. The surge has triggered widespread blocking by major platforms and raised questions about whether the web will be redesigned to serve machines rather than people. Meta's Muse agent app alone was downloaded 5 million times in its first month after launch, while apps from OpenAI and other companies have unleashed agents that operate at machine speed on behalf of individual users.

The flood of AI agents has prompted security-conscious platforms to ban them outright, the report finds. Amazon has blocked agents from Meta's Muse and filed a court appeal against AI startup Perplexity, arguing that agents threaten its advertising business by making it harder to distinguish between ads seen by humans and those served to bots. Yelp doesn't permit AI to book restaurant reservations, while Pizza Hut and other consumer sites have blocked agent access entirely, according to a TechCrunch report cited in the analysis. The vast majority of agent web traffic goes to e-commerce sites, meaning platforms have financial incentives to accommodate the new visitors. E-commerce giant Shopify is already welcoming agents by providing dedicated checkout lanes that bypass pages built for humans.

Ethan Mollick, an associate professor at the Wharton School of the University of Pennsylvania, predicts the blocking problem will soon affect the whole web. Once it becomes widely known that AI agents can request refunds or negotiate prices, "agents will overwhelm customer service departments to the point where they become unusable by real people," Mollick recently wrote on LinkedIn. He adds that "humans will eventually get very annoyed when their agents aren't allowed to do this on their behalf." The report notes that agents represent a third category of web visitor—they're not humans and they operate at machine speed, but they're also not traditional bots because they're acting on behalf of individual people to perform tasks that can be perfectly legitimate and useful.

The transformation could make the internet increasingly miserable for humans who remain, according to the analysis. Websites may become neglected as destinations for real people, with human shoppers encountering worse interfaces as platforms optimize for agents that will eventually outnumber them. Imperfect verification systems could force people to constantly prove they're human rather than agents, potentially requiring iris scans. Pricing could shift from uniform listings to different amounts displayed to different people's agents, while recommendations might be influenced by advertisements or platform commissions in ways users can't detect. The report contrasts this with promises from Google, Anthropic, and OpenAI that agents will give consumers greater power and bargaining leverage—a vision that isn't outlandish, since agents are capable of genuinely helping people by fighting bogus medical bills or researching more affordable car insurance.

Whether agents actually serve consumers depends on what rules are set for how they interact with websites and what they're allowed to do, the report concludes. These rules are still being written. Some will benefit everyone—neither consumers nor businesses want credit card information waved around by agents wherever they roam the web, and rules to keep agents more secure are a good thing, according to the analysis. But more contested rules, like whether businesses can block agents from their websites or prevent them from negotiating better deals for consumers, will determine whether the web becomes a highway for agents working for us or against us. The platforms that control access today will likely shape whether agent-driven commerce empowers individual buyers or simply creates new mechanisms for extracting value from transactions that once happened transparently. Companies building the infrastructure for this shift face a choice between designing for consumer autonomy or optimizing for their own revenue streams, with little regulatory guidance to constrain either path.