Skip to main content
Google VP Robby Stein discusses AI-powered autonomous search agents operating independently during an interview, highlighting

Editorial illustration for Google Search agents can act without user input, says VP Robby Stein

Google Search agents can act without user input, says VP...

Updated: 3 min read

Google's search engine no longer waits. According to VP Robby Stein, its new "agentic" features can act before you finish typing, fundamentally changing a tool built on passivity. The basic contract is broken: you are not conducting a search so much as supervising one.

“Ask Google to just keep you updated on anything, and now our agents can do work for you even if you're not using Google,” says Robby Stein, a vice president of product for Search.

That announcement also revealed Gemini 3.5 Flash is now the global default model for all AI Mode answers. This technical swap, along with a more responsive search box, supports the weirder shift: an engine that initiates. The practical effect is a subtle transfer of control.

The machine moves first. We react. The convenience is obvious, but it builds a habit of expecting the machine's suggestion before we even form our own question.

Google is betting its version of helpful will be enough.

Common Questions Answered

What are Google's new agentic features and how do they change the search experience?

Google's new agentic features allow the search engine to act and provide results before users finish typing, fundamentally shifting from a passive search model to an active one. According to VP Robby Stein, this means users are now supervising a search rather than conducting one, as the machine initiates actions proactively instead of waiting for complete user input.

How does Gemini 3.5 Flash impact Google Search's AI Mode functionality?

Gemini 3.5 Flash is now the global default model for all AI Mode answers in Google Search, replacing the previous default model. This technical swap, combined with a more responsive search box, enables the search engine to provide faster and more immediate responses to user queries.

What is the practical effect of Google Search's shift from passive to agentic search?

The shift from passive to agentic search represents a subtle transfer of control where the machine moves first and users react to its suggestions. This builds a habit of expecting the machine's suggestion before users even form their own question, changing the fundamental user-search relationship.

What are the potential concerns with Google's agentic search model?

While the convenience of proactive search suggestions is obvious, the agentic model creates a dependency on machine-generated suggestions that users may accept before developing their own questions or preferences. This shift in control dynamics means users are increasingly relying on Google's interpretation of what is helpful rather than driving their own search intent.

LIVE03:21OpenAI's Miles Wang in Talks for USD 2B AI Drug Discovery Startup