On Sunday night, users of Meta’s AI assistant Muse started seeing a strange error message when trying to buy items on Amazon. As discovered by GeekWire, the error message was: “Continued access by rogue AI agents violates Amazon’s Terms of Service, which you have agreed to.”
In other words, Muse is not welcome as a shopper. Anyone looking to buy something on Amazon through an AI agent will have to find it elsewhere.
It’s easy to see the bloc as a battle between big tech giants. After all, Amazon has its own cohort of underlying models, along with one of the most popular inference platforms on the internet. Unless they have a legal obligation to open the door to Muse, why would they do so?
But there are also practical reasons why Amazon may not want to get into the agent commerce business just yet. If Muse makes an incorrect order, Amazon will clean it up and deal with both the angry customer and the angry vendor. Muse has one of the lowest hallucination rates of any AI model, but it’s still quite far from zero. Even if Amazon sees an opportunity in agent-driven commerce, it may want to wait a few more release cycles before rolling out anything.
