Ring cameras and doorbells now use AI to provide specific descriptions of motion activity

Amazon-owned Ring announced on Wednesday that it’s introducing a new AI-powered feature to its doorbells and cameras, which offers users specific text descriptions of current motion activity.

Now, when users receive real-time notifications about happenings at their property, the updates will be more descriptive. For instance, “A person is walking up the steps with a black dog,” or “Two individuals are looking into a white car parked in the driveway.”

The feature aims to improve upon the vague notifications that were previously available. Now, users will be able to know exactly what is happening and can quickly decide whether it requires immediate attention. It should be noted that the AI only describes the first few seconds of the motion-activated video clip.

The feature is being rolled out today as an English-only beta for Ring Home Premium subscribers in the U.S. and Canada. Users can choose to disable the feature by going to settings in the Ring app.

According to a blog post by Jamie Siminoff, Ring’s founder and now Amazon’s VP of home security, the company plans to introduce additional AI features. One of these combines several motion events happening in and around a home into a single alert.

Ring also intends to implement a “custom anomaly alert” that allows users to define what constitutes an anomaly for their property, enabling the camera to notify them when such an event occurs. 

Additionally, Siminoff mentioned that Ring will “learn your routine,” so it can inform users when something is out of the ordinary. This may be unsettling for some users, especially considering Ring’s past privacy concerns.

“We are just starting to scratch the surface of AI. I feel like we are back to the very early days of Ring again—I see unlimited potential for new experiences we can invent for our neighbors,” Siminoff wrote. 

This announcement follows the recent launch of Ring’s AI-powered search feature, which enables users to locate specific moments within video recordings.