Who is liable? Schwab, JPMorgan, Amazon, Delta and Yelp are shutting agents out, and the Verge counted the doors
The Verge published a feature on October 17 with a blunt title: websites are closing their doors to AI agents. The reporting pulls together what users of Meta's Muse, xAI's Grok, OpenAI's Dots and Instinct's agent have been hitting over the past few weeks. Yelp only admits agents that pay into its data-licensing program. Delta blocks agents from booking tickets. JPMorgan Chase and Charles Schwab confirmed they block agents from customer accounts on security grounds. Reddit threads report Cars.com and Amtrak newly inaccessible. Amazon still blocks Muse, Grok, Instinct and Dots, telling Meta that continued access violates site policy, which prompted Musk to muse about starting a competitor to Amazon. Google is reportedly suspending some accounts that use Muse for suspicious activity, and Walmart checkout fails for some Muse users despite the Meta partnership.
The quote of the week came from Schwab's CEO Rick Wurster to Bloomberg on October 16, which the Verge ran as a short post six minutes before our window opened. We don't want someone just giving their password to Muse and letting Muse come in and take action on that client's behalf. We can't control it, the client can't control it, something could go wrong and who's liable? So we're not allowing it. Bloomberg's framing was that US banks are blocking agents attempting to move money for customers. That is the agent-authority gap stated by the person who carries the liability, and it is the same gap Amex tried to paper over this week with its Intent Intelligence token and Shopify tried to route around by opening its catalog to any agent.
The Verge's analyst voice was EquiStamp CEO Chris Canal, who said the blocks absolutely affect the perceived value of the Anthropics and the OpenAIs, and pointed out that anti-bot systems built for scammers fire on agents too, because an agent filling a form at machine speed looks exactly like the thing those systems were built to stop. Meta's response is partnerships, Walmart, Gap, Sephora and Best Buy, plus the argument from its own post that turning away a personal agent means turning away the customer. The retailers' response so far is that a customer who hands a password to a bot is not a customer they can protect.
Put the week's pieces in a row and the shape is clear. Yelp wants to be paid for access. Banks want a liability framework before they let anything move money. Amazon wants agents to be its agents. Shopify and Amex are building the credential layer that would let a merchant tell a legitimate delegated purchase from a scripted scrape. The agent companies are shipping the agents first and negotiating the doors second, and this feature is the first mainstream count of how many doors that is.
Links: https://www.theverge.com/ai-artificial-intelligence/1013195/ai-agents-blocked-web , https://www.theverge.com/ai-artificial-intelligence/1013303/charles-schwab-is-blocking-ai-agents-too
← Back to all articles
The quote of the week came from Schwab's CEO Rick Wurster to Bloomberg on October 16, which the Verge ran as a short post six minutes before our window opened. We don't want someone just giving their password to Muse and letting Muse come in and take action on that client's behalf. We can't control it, the client can't control it, something could go wrong and who's liable? So we're not allowing it. Bloomberg's framing was that US banks are blocking agents attempting to move money for customers. That is the agent-authority gap stated by the person who carries the liability, and it is the same gap Amex tried to paper over this week with its Intent Intelligence token and Shopify tried to route around by opening its catalog to any agent.
The Verge's analyst voice was EquiStamp CEO Chris Canal, who said the blocks absolutely affect the perceived value of the Anthropics and the OpenAIs, and pointed out that anti-bot systems built for scammers fire on agents too, because an agent filling a form at machine speed looks exactly like the thing those systems were built to stop. Meta's response is partnerships, Walmart, Gap, Sephora and Best Buy, plus the argument from its own post that turning away a personal agent means turning away the customer. The retailers' response so far is that a customer who hands a password to a bot is not a customer they can protect.
Put the week's pieces in a row and the shape is clear. Yelp wants to be paid for access. Banks want a liability framework before they let anything move money. Amazon wants agents to be its agents. Shopify and Amex are building the credential layer that would let a merchant tell a legitimate delegated purchase from a scripted scrape. The agent companies are shipping the agents first and negotiating the doors second, and this feature is the first mainstream count of how many doors that is.
Links: https://www.theverge.com/ai-artificial-intelligence/1013195/ai-agents-blocked-web , https://www.theverge.com/ai-artificial-intelligence/1013303/charles-schwab-is-blocking-ai-agents-too
Comments