The list was created as part of Microsoft’s Computer Science Education Week celebrations. The company chose to reflect its desire for everyone to be digitally literate by asking some of its female researchers to share their predictions. While women and girls represent around 50 percent of the world’s population, they form just 20 percent of computer science graduates.
A frequently cited reason for this divide is the lack of female role models for girls interested in STEM subjects. In an attempt to counteract this, Microsoft asked 17 women working in its global research departments for their opinions on what’ll change in technology between 2017 and 2027.
One of the most intriguing developments was suggested by distinguished scientist Susan Dumais. According to Dumais, the search box “will disappear” from interfaces in the next ten years. It will be replaced by automatic search functionality that’s more “ubiquitous, embedded and contextually sensitive.” This is already becoming evident as devices like the Amazon Echo and Google Home let you start a voice search in your home through barely any effort.
Future search interfaces will also handle the context of your query, determining meaning based on your location, the people with you and the activity you’re performing. Furthermore, additional search types will see more use, including search using images, sounds and video as the source.
According to one of the researchers, computers will learn to “see” during the ten-year period. Xiaoyan Sun suggested that imaging devices and deep learning techniques will become so sophisticated that artificial “eyes” could be created. These would enable computers to assist humans in a variety of roles and professions, including medical and security applications.
Further developments in deep learning will come from the creation of fair and accountable algorithms, distinguished scientist Jennifer Chayes said. By 2027, researchers will have developed algorithms that can avoid biases and manipulation, creating more robust systems. Current deep learning mechanisms often create the same data discriminations as humans do, a sign of their continued reliance on human input.
In a different field of AI, the strides forward in cognitive computing will allow next-generation natural language processing systems to communicate with humans “relatively effortlessly.” Microsoft’s Kalika Bali said she expects computers to start negotiating and debating with humans. They’ll be able to engage in social situations and offer persuasive powers, creating culturally aware AI.
The full list of predictions can be viewed over at Microsoft’s official blog. It includes statements on the future of mobile computing and the evolution of biological computation in 2017 and beyond. The overarching message from the researchers is that developments in machine learning will enable the creation of computers that feel and act like humans.
We’ll get even closer to building true artificial intelligence and machines will help to solve significant societal issues by impartially reasoning. While some of the predictions may seem far-fetched today, just ten years ago the iPhone was yet to launch and virtual reality was little more than a concept. Given another decade, technology could undergo another revolution, either based on or alongside Microsoft’s predictions today.
