Among the featured speakers at the O’Reilly AI Conference were Lili Cheng of Microsoft Research, and Oren Etzioni, director of the Allen Institute for Artificial Intelligence. Their views offer contrasting perspectives: AI product design and development lessons from Microsoft and some hard data about expectations for market development from the AI think tank.
Lili Cheng, Microsoft Research. Microsoft’s bot, Xaoice, now with 40 million users, was rolled out first in China, and then Japan. Like Amazon’s Echo, it has a suite of engaging, amusing, and useful characteristics that keep people coming back. Unlike the Echo, it is stateful, and conversational. In addition to just chatting for fun, it can translate, schedule meetings, answer math problems (and joke about them), recognize dogs, and count sheep for you when you can’t go to sleep. Moving Xaoice to the US as Tay.ai was, however, a disaster. Cultural hurdles as well as back doors in the software can send a bot off the culturally permissible rails. Microsoft has taken the technology and released an open source bot framework: https://dev.botframework.com. A high hurdle to jump: building a cross channel bot across walled gardens. This has been one of the major challenges for cognitive computing: automating the ingestion of data from multiple sources in different formats, with differing terminology.
Oren Etzioni, Allen Institute for Artificial Intelligence: “Machine learning today is 99% human work.” “Superhuman performance on a narrow task is not human performance.” Etzioni’s cold splash of reality is important. AI has been so subjected to hype cycles that we all have unrealistic expectations about what it can accomplish. Right now, the reality is that machine learning, conversational systems, bots, etc., are a boon to humans trying to navigate too much information within a well defined, narrow field. Will they eventually take over the world? Etzioni’s research in the field indicates that AI researchers themselves think otherwise. He asked them: “When will AI (superintelligence) arrive?” 67% answered that it will take more than 25 years. 25% said that it will “never” arrive. That’s a comforting result. Humans need all the help we can get in making sense of complex information landscapes. But it sounds like the final say is still up to us.Share