Who's Afraid of AI?

For decades, the pioneers of aviation promised grandiose feats, only to fall short again and again. But then the Wright brothers had a breakthrough—their first flight in Kitty Hawk, North Carolina—and the technology took off. And so it is for AI today: After many years of relatively slow, underwhelming progress, the technology is finally starting to perform, and now a cascade of breakthroughs are flooding the market, with many more in the works.

As the old saying in AI research goes, what’s hard for people is easy for machines, and vice versa. Driving a car, which involves thousands of small but nevertheless complex decision scenarios during each trip, was previously impossible for computers. Why is that changing now? Because software that learns from data in connection with controllable hardware has increasingly mastered three core skills—recognition, insight, and implementation of an action.—

Whether machines can think will probably remain a question that philosophers discuss for decades longer. But an intelligent machine that passes the test set by Alan Turing is probably only a matter of a few years away.

The essential difference between artificial intelligence and classic IT systems lies in this ability of AI to independently improve its own calculations by classifying the results of its actions.
Auto-correction is built into the system.

When AI finds its way into medical practices, a fundamental question arises: Do we better trust the judgment of an artificial neural network based on data than that of an experienced doctor who might have been treating us since childhood? Maybe a computer-science student would answer with an unconditional yes. That student believes in statistics. Some patients will struggle with transferring the decision-making power from a human being to a machine.

Cultural attitudes speed up or slow down the acceptance of innovations. Robots are enemies in Europe, servants in America, colleagues in China, and friends in Japan.

Artificially intelligent systems will not enslave humanity in the foreseeable future. Nobody knows what computers will be able to do in two hundred years, but what we know for now is that computer scientists don’t know any technological path that could lead to artificial superintelligence.
The end of the world has been postponed once again

We can delegate decision-making to machines in many individual fields. AI systems that are skillfully programmed and fed the proper data are useful experts within narrow specialties. But they lack the ability to see the big picture. The important decisions, including the decision about how much machine assistance is appropriate, remain human ones. Artificial intelligence cannot relieve us of the burden to think.

"Exciting, mind boggling, and, yes, unsettling, AI is destined to transform nearly all sectors of industry and our lives in ways that we can’t even imagine today. At an inviting length, Who’s Afraid of AI? is a wonderful primer on what we need to know and what we need to worry about.”

Foreword Reviews