Get the newsletter
Google · AI / ML · Filed Oct 29, 2025 · USPTO published Oct 1, 2026 · Oct 1 release

Google Patents a System That Turns Smart Home Sensor Data Into Suggested Actions

Your smart home collects a mountain of data every day, but most of it just sits there. Google is patenting a way to make that data work, using an AI model that spots patterns across your sensors and turns them into concrete suggestions.

A smart home with various sensors and connected devices, including a delivery person and a dog with a smart collar. Drawing from patent filing US 2026/0303401 A1.
A smart home with various sensors and connected devices, including a delivery person and a dog with a smart collar.All 13 drawings ↓
Publication number
US 2026/0303401 A1(US20260303401A1)
Applicant
Google LLC
Filed · Published
Oct 29, 2025 · Oct 1, 2026
Inventors
Ignacio Robles Paiz, Raymond Stepkans Strods, Erin Rebecca Griffiths, Adam Cutbill, Suneil Kamireddy
Status when we published
Waiting for an examiner
Document
20 claims
US classification
700/275
Parent application
Claims priority from a provisional application 63780826 (filed Mar 31, 2025)

What Google's home-pattern AI actually does for you

Every time your front door opens, your thermostat adjusts, and your lights flip on, your smart home logs all of it. That data piles up fast, and right now most people never see any value from it beyond the individual device doing its one job.

Google's patent describes an AI that watches all those logged events together, finds patterns across them, and then proposes specific actions based on what it learns. So if your motion sensor and thermostat data consistently show that you're home by 6 p.m. on weekdays, the system could suggest an automation to pre-heat the house before you arrive. You'd get a suggestion to act on, rather than having to dig through settings yourself.

The idea is to close the gap between owning a smart home and actually getting the most out of it. Most people set up a handful of routines at install and never change them. This system is meant to notice what you've been missing.

From the filing · CLAIM 1
generating, by one or more home surveillance sensors, home data; receiving, by a machine-learned (ML) model, the home data; generating, by the ML model, one or more correlations based on the home data …

Translation: Sensors collect activity data and an AI model analyzes it to find patterns.

How the ML model connects sensor events to actions

The patent describes a pipeline that starts with home surveillance sensors (motion detectors, door/window sensors, cameras, thermostats, and similar devices) continuously generating what the filing calls home data, a stream of timestamped events from around the house.

That data is fed into a machine-learned (ML) model, which finds correlations (repeating relationships between events, like the porch light turning on shortly before the front door opens every evening). The model doesn't just log those patterns; it uses them to generate a home action, a specific proposed change to how the home operates.

The action is then configured for output to the user, meaning the system presents it in a form the person can accept or reject rather than applying it automatically without consent. The patent doesn't prescribe exactly how that suggestion reaches the user, whether as a push notification, a card in an app, or a voice prompt, but the output step is explicitly part of the claimed method.

  • Sensors generate continuous event logs
  • ML model finds repeating cross-device patterns
  • System proposes a concrete home action based on those patterns
  • User receives the suggestion for review

What this means for Google Nest and Home users

For most people, the hardest part of a smart home is the setup, specifically knowing which automations to create. You have to think of the scenario yourself, find the right settings, and configure it manually. A system that watches your actual behavior and proposes automations removes that friction entirely. You'd notice it most the first time it suggests something you'd been meaning to set up for months.

Google's interest in ambient home intelligence shows up clearly here. The patent also has a privacy dimension worth noting: the model processes data from surveillance sensors, so how that data is stored, who can see it, and whether it leaves the device are questions this filing doesn't fully answer. Those are the details users will reasonably want before trusting a system like this.

Google's 46th filing we've tracked since May in our AI agents acting for you watch builds on earlier applications like formula writing from examples and proactive on-device suggestions.

Source: US 2026/0303401 A1 at the USPTO (PDF). Full patent text and figures are in the official publication.

Get our take in your Top Stories

Liked this breakdown? Add Patentlyze as a preferred source on Google, and our plain-English take shows up more often in your Top Stories the next time Google patent news breaks.

There are more where this came from

We read every patent application Big Tech publishes and send you the ones worth knowing. Plain English, free, every week.

The drawings · 13 sheets

13 drawing sheets from US 2026/0303401 A1 · click any drawing to enlarge

Smart home environment with sensors., drawing sheet 1 from US 2026/0303401 A1
Sheet 1 Smart home environment with sensors.
People interacting with smart devices., drawing sheet 2 from US 2026/0303401 A1
Sheet 2 People interacting with smart devices.
Various smart devices and their components., drawing sheet 3 from US 2026/0303401 A1
Sheet 3 Various smart devices and their components.
Phone screen showing home insights., drawing sheet 4 from US 2026/0303401 A1
Sheet 4 Phone screen showing home insights.
Phone screen showing suggested actions., drawing sheet 5 from US 2026/0303401 A1
Sheet 5 Phone screen showing suggested actions.
Smart home system data flow., drawing sheet 6 from US 2026/0303401 A1
Sheet 6 Smart home system data flow.
Machine learning model generation., drawing sheet 7 from US 2026/0303401 A1
Sheet 7 Machine learning model generation.
Text processing for machine learning., drawing sheet 8 from US 2026/0303401 A1
Sheet 8 Text processing for machine learning.
Show the other 5 sheets
Google Patents a System That Turns Smart Home Sensor Data Into Suggested Actions patent drawing, sheet 9
Sheet 9
Google Patents a System That Turns Smart Home Sensor Data Into Suggested Actions patent drawing, sheet 10
Sheet 10
Google Patents a System That Turns Smart Home Sensor Data Into Suggested Actions patent drawing, sheet 11
Sheet 11
Google Patents a System That Turns Smart Home Sensor Data Into Suggested Actions patent drawing, sheet 12
Sheet 12
Google Patents a System That Turns Smart Home Sensor Data Into Suggested Actions patent drawing, sheet 13
Sheet 13
Patent filing page

Watchlists this filing belongs to

All watchlists →
Reader comments

No comments yet

Start the discussion

Real name or a handle, either is fine. Comments are read by a person before they appear, so allow a little time. Keep it about the filing.