With the success of molecular biology explaining the mechanisms of life we have lost sight of the question one level up. We do not have any good answers at a more systems level of what it takes for something to be alive. We can list general necessities for a system to be alive, but we can not predict whether a given configuration of molecules will be alive or not. As evidence that we really do not understand what it takes for something to be alive, we have not been able to build machines that are alive.
Everything else that we understand leads to machines that capitalize on that understanding — machines that fly, machines that run, machines that calculate, machines that make polymers, machines that communicate, machines that listen, machines that play games. We have not built any machines that live.
RODNEY A. BROOKS is director of the MIT Artificial Intelligence Laboratory and Chairman of iRobot Corporation. He builds robots.