Although we live in a connected world where sharing information is as easy as touching a screen on a smartphone, people who are blind are unable to access much of the technology that the sighted take for granted. Facebook’s Accessibility team is trying to change this by using AI to help people who are blind share information on its platform. Many people who are blind interact with computers and smart devices by using navigation software called “screen readers” to turn written digital content into speech output. But because the software cannot “read” pictures, the content it can access is limited to text. To help those with visual impairments experience pictures, Facebook has developed a way to make images visible to machines by using AI and neural networks. The technology works by recognizing the image in a picture and providing its description in a form that the screen reader can translate. According to Facebook, its object recognition technology is now able to recognize over 80 objects and activities, and the system is expected to improve with more training. The longer-term goal is for the technology to be able to recognize human faces. Matt King, who lost his sight in college due to retinitis pigmentosa, is the engineer who led the Facebook Accessibility team in developing this technology. He acknowledged that there is much room for improvement in terms of creating technologies that serve those who are blind or visually impaired, especially since the so much of the information available on the Internet is in visual form. The project is already benefiting around 50,000 people. However, the World Health Organization estimates that there are 39 million people who are blind and 246 million people who are partially sighted worldwide. Many of these people’s lives could be dramatically impacted by Facebook’s object recognition technology, as well as by future applications that rely on AI technologies.