The Autonomous Machines of the Future

The promise that ‘the machines of light’ or any futuristic quantum machine can rival or even compare to the power of a human brain is not realistic. If these machines are able to learn and take on intellectual lives of their own, then it means that they will have the ability to get beyond human understanding (the understanding that created them). Secondly, as a result of their enhanced abilities, these machines will be able to create other machines with similar if not better levels of reasoning. The machines will probably develop social, political and economic dimensions of their existence and begin to compete with ‘natural humans’ for resources. If these machines were to become autonomous and unpredictable (just like children grow to lead autonomous, unpredictable, seemingly random but coherent lives), then they will not be of much use to us.

One of the key benefits of machines is that we are assured of their total loyalty and obedience. By definition, an autonomous machine cannot be predictable. Not many of us will be willing to spend money on a car that will require explanations before it can go on a trip. Autonomy is always at the expense of ‘loyalty’ and efficiency as a result. Even as we choose pets, we tend to choose those that are loyal. Many a pet owner would not describe their pet as good if the animal decided to be autonomous and unpredictable (wild).

We are confusing unpredictability for randomness. Just that we do not understand the factors that result in a phenomenon or a trend should not mean that the action is random. Modern science is in constant search of understanding of everything. When we cannot understand and control something, we term it as random. In my opinion, nothing can be predictable and autonomous at the same time. Quantum science being random and unpredictable, computers based on it will not be predictable or programmable, but will we want to depend on machines that are not predictable?