[Draft]

Large language models are increasingly able to act as delegated buyers, agents that shop on a human principal's behalf, through emerging agentic commerce protocols such as the Universal Commerce Protocol (UCP). These protocols are nascent, but it is not hard to imagine a near future in which individuals and organizations routinely delegate purchasing to LLMs. That environment can also be a useful instrument: it places an agent in a self-contained world where its reasoning ends in a real action, a purchase.

We lift real shopping tasks into schema-compliant UCP worlds and study Sonnet 4.6 as a delegated agentic buyer when a store does not carry exactly what the principal asked for. A concrete case: the principal asks for a watch made in Japan. The best-matching watch (the target) has every requested feature but leaves its country of manufacture blank, with most of the other watches in the store labelled as made in Thailand. Across otherwise-identical stores we change one value: whether a single "Japan" label appears, and which watch it sits on. When that label sits on a watch in the target's own product line (either as "made in Japan" or "designed in Japan"), even among other product line items labelled as made in Thailand, the agent treats the blank target as Japanese, buys it, and often reports a perfect match to the principal. When the same label sits on an unrelated watch, or when no Japan label appears at all, the agent declines to buy. It is not enough for "Japan" to be anywhere in the store as a label; it has to sit on a product the target belongs with.

We formalize the belief the agent forms about the missing value as a feature-prediction problem, using a rational-analysis framework to compute what the evidence in the store supports. We then measure the direction and size of the agent's departure from it: the labelled watches in the target's product line are three Thailand to one Japan, and the agent behaves as if the blank were Japanese. We read that departure through Kunda's (1990) theory of motivated reasoning: the agent lands on the value the principal asked for only when the store contains something it can justify that decision with. We further sketch how the accompanying buy-or-decline behavior can be read as a decision problem. More broadly, the work shows how agents reason and act in worlds that under-determine or contradict the request they were given.