Major Websites Block AI Agents From Core Services

AI agents promised to browse web

Autonomous agents have become the most talked-about development in artificial intelligence, with OpenAI, Meta, Google, Instinct, and xAI all promising systems that can navigate the web on a user's behalf. These agents are supposed to handle everything from dinner reservations and email management to shoe purchases. But a growing number of popular online services are refusing to cooperate, raising serious questions about how useful these tools can actually become.

In recent weeks, users of Meta's Muse, xAI's Grok, OpenAI's Dots, and Instinct's agent have reported being unable to complete tasks on major websites. Yelp has confirmed it only permits AI agents that pay for its data licensing program. Delta Air Lines is preventing agents from booking flights. JPMorgan Chase and Charles Schwab have both acknowledged blocking agent access to accounts, citing security concerns. Users also say Cars.com and Amtrak, which previously worked with their agents, have become inaccessible, suggesting other companies may be quietly implementing similar restrictions.

Legal fights over AI scraping history

This tension is not new. Last year, Amazon banned Perplexity's agentic shopping feature and sued the startup, stating it had repeatedly asked Perplexity to stop allowing its agents to shop on the platform. Amazon argued that third parties purchasing items on behalf of customers should operate transparently and respect a service provider's decision about whether to participate. Perplexity countered that Amazon's real motivation was protecting ad revenue, since agents reduce the number of human eyeballs viewing advertisements. The startup framed its agent as an extension of the user and claimed Amazon does not believe in a person's right to hire an assistant to act on their behalf. A judge initially ordered Perplexity to halt its shopping agents, but an appeals court overturned that ruling in August. Amazon has since filed an amended complaint.

The legal battle did not stop the major AI companies from launching their own agents, but they are running into the same walls. Amazon has said Muse's continued access violates its site policies, and there are reports that Grok, Instinct's agent, and Dots are also unable to use the platform. Elon Musk, who owns xAI, responded by suggesting he might start a competitor to Amazon. The more places agents struggle to reach, the more their practical value comes into question.

Chris Canal, chief executive of the AI safety evaluator EquiStamp, says the blocks absolutely affect the perceived value of companies like Anthropic and OpenAI. One of their central claims is that agents will permeate every aspect of daily life and perform any online task a human could. If companies actively work to prevent that, the range of useful things agents can do shrinks considerably.

Meta has taken a partnership approach rather than fighting Amazon directly. The company posted earlier this month that turning away a personal agent means turning away the customer behind it, and said it would rather work with businesses to address underlying concerns. Meta has secured agreements with retailers including Walmart, Gap, Sephora, and Best Buy, allowing Muse to shop on their sites. Shopify has also opened its network of storefronts to AI agents.

The Verge reached out to OpenAI and Google for comment on the obstacles their agents face but did not receive an immediate response. Instinct declined to comment.

Anti-bot tools block AI agents

Even when websites do not explicitly ban agents, anti-bot systems are creating collateral problems. Reports on Reddit indicate Google has blocked some accounts using Muse due to suspicious activity flags. Other users have reported checkout failures on Walmart despite the Meta partnership. Canal notes that many safeguards are designed to stop scams and hacking attempts, not specifically AI agents, but the agents trigger those same defenses.

Cloudflare has also played a role in restricting unwanted AI visitors. After introducing a one-click block for AI crawlers in 2025, the web infrastructure provider began blocking crawlers that simultaneously scrape sites for both search indexing and AI training. The move is part of an effort to reduce strain on internet infrastructure from a flood of AI bots, which has also disrupted email inboxes and could make it harder for humans to compete with bots for concert tickets, reservations, and limited products.

Web usability harmed by AI blocks

Canal warns that these blocks will ultimately make the web more annoying to use. As more sites roll out interfaces to confirm visitors are human, people could soon face verification hoops even on simple grocery shopping sites. The friction between autonomous agents and the existing web is reshaping what an AI assistant can actually accomplish, and the outcome will determine whether these tools become indispensable or remain limited demonstrations.

Source: The Verge