Artificial Intelligence is Here: Siri, Cortana and Google Now
Computers can now do things that even just a few years ago were considered solely the stuff of science fiction. And, as in so many other areas, the tech titans – Apple, Microsoft and Google – are fighting tooth and nail to dominate the newest high-growth, transformative areas of tech. One such area is artificial intelligence (AI) powered personal assistants. All three companies have software that now competes in this space. Apple has introduced the Siri personal assistant, Microsoft has Cortana and Google offers Google Now. The technology is still in its early stages but it is already exciting to see glimpses of a very dramatic future where computers are likely to be much smaller and lighter than they are now and where people can interact with them without need of a keyboard or a screen. Computers embedded in clothing and other wearables – especially those with an active Internet connection – will make some pretty amazing things possible in just a few years.
DARPA and the Future of AI
While all of this AI personal assistant business feels pretty cutting-edge, it is actually a long time coming. Most people of a certain age can remember the computer HAL from 2001, A Space Odyssey. People have been dreaming about AI assistants for a very long time! Apple’s Siri system has been out since 2010 and the predecessor company was acquired by Apple a few years before that. The modern basis of AI personal assistants was established, perhaps surprisingly, in a program funded by the US Department of Defense. The Defense Advanced Research Projects Agency (DARPA), with an annual budget of almost $3 billion, is tasked with carrying out important technological development projects that could make a contribution to national defense. About 15 years ago, DARPA funded a project called PAL – the “personalized Assitant that Learns.” The PAL system was intended to be able to handle automated data retrieval and data synthesis – essentially taking the search engine to an entirely new level. Two of the key concepts of the PAL program that have carried over into all of the current AI personal assistants are first, that the software should learn over time to be able to access Internet content and then put it together I useful ways for the user and, second, that the software should actually learn from the user and become better and better able to anticipate his or her needs and respond to them.
The AI Personal Assistant and the Future
In thinking about the future of mobile computing, what is truly exciting is to imagine how the software and the hardware are evolving together to make powerful applications possible. While computing power continues to increase geometrically and while the size and weight required for the computer power drops, the weak link has always been the interface between human and machine. Humans have typically interacted with computers by typing their instructions and reading the responses from a screen. Because of limitations of human physiology, computer relying on this kind of interface can only be made so small. But microphones and speakers have already been miniaturized to a great degree. Now with software that makes oral/auditory interface possible, the size limitations are resolved. Not only is this great news for the vast majority of computer users, it is also good news for the visually and physically impaired who in the future will be able to interact with computers just like anybody else. When you add into the scenario universal Internet access and software that is increasingly powerful at anticipating the information required by a user, collecting the useful information from an ever broader, deeper Internet and synthesizing the information in useful ways, the results of that kind of technology are truly amazing.
The things that today’s AI personal assistant can do are just the tip of the iceberg. In just a very few short years, true Internet-enabled AI will be widely available from your eyeglasses, your wristwatch or your sport coat. Get ready for a wild ride!