The Doppler Quarterly Fall 2017 - Page 33

objective of much of the AI-driven automation is to remove people from the process where possible, because human interaction causes latency, inaccu- racy and avoidable errors. Furthermore, if we want or need to talk to our com- puter—that is, where a person truly needs to be part of the dialog—we have a keyboard and a mouse. And when using them, we won’t bother anyone else in the room. You can’t say the same for voice systems. Nor do you have to worry about a voice system respond- ing out of line because it hears something not meant for it—a common issue with Siri, Cortana, Google Now, Alexa, and their ilk. On the other hand, chatbots have really been devel- oping as a way to get automated responses for chat- based interaction, such as Skype, Slack or Facebook Messenger. Replacing the h uman side of that interac- tion with voice started in the world of interactive voice response (IVR) so we could get automatic responses via phone without being limited to the phone’s sparse keyboard. More recently though, we’ve seen the emergence of more device interaction via Siri, and now Alexa. Many people spend a lot of time in the car and in order to avoid tickets or injury they’ve made it a point to really stay hands free, while still constantly using an iPhone. We’ve learned to interact with our devices via keyboard and mouse ONLY because that was pretty much our only choice for the past 4 decades. Conceptually it would be much faster if we could use a combination of voice and gestures instead. The technology is not quite perfected yet, but it is improv- ing dramatically as we’ve started to apply better deep learning under the hood. Imagine when we realize the concepts presented in the movie Minority Report - virtual big screen pro- jected from a small handheld device like a phone with interaction via voice and gesture. Untethered and interacting closer to the speed of thought. Another cool concept that emerged in the recent HBO series Westworld was the idea that eventually human programmers will reach the limit of what they can do. So they built true AI that could start to build out the code base faster and more efficiently. Replac- ing programmers with AI? Now you’re talking! Won- der if our stand-ups will get any faster? FALL 2017 | THE DOPPLER | 31