GoKawiilGoogle has a massive, constantly refreshed "shopping graph" that tracks more than 60 billion product listings, and it wants to turn its AI on that giant catalog to help you find the right product for the right price.
At its Google I/O annual developers conference Tuesday, the company revealed a handful of new features that aim to turn a typical Google search into a personal shopper experience driven by agentic AI.
Suresh Ganapathy, Google's senior director of consumer shopping, told reporters ahead of I/O that he hopes the AI tools create shopping experiences that feel "fun and powerful and intelligent."
Letting AI do the hard part of shopping
The announcements at I/O focus on making agent-driven commerce seamless across discovery, purchase and post-purchase tasks.
"On the shopping side, I'm really looking forward to a world where shopping feels really fun," Ganapathy said. "We keep hearing from shoppers that they really enjoy the fun aspects of shopping, but would love to delegate the more tedious parts to AI."
To handle the hard parts, Google is using UCP, a shared language co-developed with major retailers and platforms (Shopify, Wayfair, Target, Walmart, etc.). UCP is described as letting agents and systems operate together across a shopper's journey.
Another AI shopping tool, Agentic Payments Protocol -- or AP2 -- allows agents to buy things under user-defined constraints.
"Imagine you could go tell your agent to buy something within certain budget constraints from a certain set of merchants and your agent is able to go do that, and it only buys it if it meets all of those criteria, and it keeps your data safe and secure," Ganapathy said.
The protocol protects payment data through digitally signed contracts that only disclose information to necessary parties. Google plans to bring AP2-based products to its platform this fall.
... continue reading