Science fiction has promised us a whole lot of technology that it's rudely failed to deliver--jetpacks, flying cars, teleportation. The most useful one might be the robot companion, à la Rosie from The Jetsons, a machine that watches over the home. It seemed like 2018 was going to be the year when robots made a big leap in that direction. Two machines in particular surfaced to much fanfare: Kuri, an adorable R2D2 analog that can follow you around and take pictures of your dinner parties, and Jibo, a desktop robot with a screen for a face that works a bit like Alexa, only it can dance. But then, as quickly as the home robots came, they disappeared.

This is a guest post. The views expressed here are solely those of the authors and do not represent positions of IEEE Spectrum or the IEEE. We started TickTock in March 2017, knowing that robotics was about to have a breakthrough, and it was going to start with mobility. Soohyun Bae and I both met at Google years earlier, and had worked on augmented reality products. Soohyun went on to Magic Leap, and I helped launch Project Tango, now called AR Core. We both knew that AR's push for mobile 3D mapping and scene understanding was causing a dramatic tech shift that would also benefit robotics.

TickTock has run out of time. Don't fret if you don't know what that is--after all, the startup launched just a year ago. After TickTock's collapse, though, co-founder and ex-Googler Ryan Hickman is talking candidly about what it's like to build an unwanted robot. Well, a robot unwanted at least by venture capitalists--some 200 investors that TickTock tried to convince to cough up cash before the startup closed down. But with the demise of TickTock come valuable insights into the robotic home of the future, and which companies will end up conquering it.

We describe TickTock, a conversational agent designed to engage humans on topics of its choosing and to carry on an interaction for as long as possible. Our prototype uses a database of talk show transcripts featuring guests from the film industry. To be an interesting companion Tick Tock uses immediate context from the last two turns to formulate queries into a database of utterances. The process is automatic. TickTock monitors user engagement and performs certain moves, such as topic shifts, based on its assessment of user state. Initially we used utterance content for monitoring and subsequently we begun to investigate non-language cues, such as prosody and visual cues to create a more robust engagement model based on multiple human communication channels.