You Can’t Chit-Chat With a Chatbot

You Can’t Chit-Chat With a Chatbot

David Bruemmer, the co-founder of 5D Robotics, is pretty optimistic about the robot revolution, believing that robots will become our peers, not our replacements, and create more good jobs than the low-paying, menial jobs they destroy. Writing at TechCrunch, Bruemmer recalls a conversation with his hairdresser that got him thinking about the intangible, creative, human aspects of some jobs that he doubts robots will ever be able to take over:

I seem to have very interesting discussions whenever I get a haircut. One day I was explaining to my hairdresser that I work with robots and she stopped cutting, visibly upset and remarked that she would probably be out of a job before too long because of the development of robotic intelligence. I thought about it, and realized that robots would most likely replace my job before they replaced hers.

Her job was really, really hard. She had to take hopelessly vague tasking like: “Give me a George Clooney haircut but kind of a bit more fun and crazy.” What does that mean? I can pretty well guarantee that a robot won’t ever know what that means. Regardless of trying to understand the semantics, a much harder task is figuring out how many small scissor cuts will result in an emerging look. Cutting hair is not an easy job; I know I could never do it. I believe it would scare any robot as well.

I told my hairdresser she had nothing to worry about, but she still looked at me skeptically. What she didn’t realize is that even if robots could cut hair effectively, she still had a major advantage. She had softly massaged my head and neck before she started cutting and she breezily discussed just about any topic while making her customer feel at ease. Machines will impact the jobs of surgeons and professors long before they impact her job.

In other words, Bruemmer realized that his hairdresser’s job requires making a uniquely human connection with her customers, both in creatively interpreting their vague instructions and in making casual conversation to keep them relaxed while she works with sharp objects in close proximity to their heads. That type of conversational skill is one human quality robots are nowhere near acquiring, at least not yet, futurist Liz Alexander argues at Fast Company, and they won’t be able to completely replace human beings at many jobs unless or until they do:

[I]n order for artificial intelligence to take a much bigger bite out of the knowledge-economy workforce, the technology may need to start behaving more like humans, not less. And that will mean mastering one key behavior: small talk.

Sociolinguists involved in the Language in the Workplace Project at Victoria University of Wellington in New Zealand have discovered that people switch naturally between “transactional” talk—such as discussing a business goal—and “interactional” talk, like when you encourage or show concern for a distressed coworker. We do this toggling back and forth with our colleagues all the time without even realizing it. It’s at the heart of how we communicate.

So far, this kind of mental flexibility isn’t something machines—designed to execute one type of task, consistently and flawlessly—can currently manage. And they won’t, according to machine-learning expert Geoffrey Hinton, until machines’ neural networks of 1 billion synapses come closer to the 1,000 trillion synapses of the human brain.