The Most Human Human

I really enjoyed this book. The framework is that the author wants to win the Most Human Human award at the annual Loebner Prize, a competition that is actually focused on the AI community. In preparation for this event he decides to investigate what it actually means to be human, with his research, conversations, and introspections comprising the actual bulk of the book. This idea - what it means to be human - takes the reader on a free-ranging journey, so much so that at times I actually stopped to think, “Wait, what is this book about?” Once I remembered the overall subject matter, the topics would make sense in the book; even though the author talks about things that seem unrelated, they all fit nicely together in the context of how they make us human (or less-than-human, which is more interesting). By the final chapter I was getting a little impatient for the end, which the author actually discusses in general at one point. Overall, very enjoyable.

The AI community meets together to assess the “human-ness” of the latest and greatest in computer technology. The best computer is awarded “the Most Human Computer” award. And, in addition, there is a small additional contest for “the Most Human Human.”

Brian Christian decides to compete for “the Most Human Human.” He is an intriguing person who has degrees in computer technology, philosophy, and poetry. (That is such an odd combination of degrees that I feel led to repeat it again: He is an intriguing person who has degrees in computer technology, philosophy, and poetry.)

Christian thinks about what it means to be humaan in the world in which we live, the world that was, and the world that is to come.

I had many ah-ha moments while reading this book:

Christian relates a story about an author who attempts to outsource his customer service department only to be besieged by new reps calling to ask how to solve problems. In desperation, the author finally sends the reps an e-mail that says, “Don't ask me for permission. Do what you think is right.” Suddenly, the calls to the author cease and customer service ratings rapidly improve.

“Micromanagement; the kaizen-less assembly line; the overstandardization of brittle procedures and protocols...these problems are precisely the same problem, and pose precisely the same danger, as does AI. In all four cases, a robot will be doing your job. The only difference is that in the first three, the robot will be you.”

“...kids are much quicker studies at learning to ski because they are not afraid to fall. Fail and recover.”