The social network announced the update in a blog post on Thursday. Joaquin Candela, Facebook’s director of applied machine learning, wrote that the new AI-based image tool would recognize photos at the “pixel level.” So users would be able to get results based on the content of photos, even when they’re not manually tagged.

“Until recently, online search has always been a text-driven technology, even when searching through images,” Candela wrote. “Whether an image was discoverable was dependent on whether it was sufficiently tagged or had the right caption — until now.”

The new search tool will also help the network to better identify inappropriate content and spam, according to Candela.

Next, Candela would like to apply this “image understanding” to Facebook’s automatic alternative text—a feature developed to describe the content of photos to the visually impaired.

“Until recently, these captions described only the objects in the photo,” he wrote. “Today we’re announcing that we’ve added a set of 12 actions, so image descriptions will now include things like ‘people walking,’ ‘people dancing,’ ‘people riding horses,’ ‘people playing instruments,’ and more.”