Intelligence, in general terms, is the ability to collect and correctly apply information to a situation. Artificial intelligence, then, is the ability of a computer to return an accurate solution, which is not found in its stored data, to a query.
Artificial intelligence problems are those queries, situations, evaluations, etc., which involve more variables and complexity than a human can accurately remember and compute. Many refer to these problems as “NP” or “intractable”. A good example of such a problem is in the creation of medicine. Imagine that researchers set out to create a cure for the common cold. Their research identifies one hundred different chemicals that possess specific properties that might, through some combination with each other, produce the desired effect. For the sake of conversation, let us give the researchers some help and say that each chemical has only one property. While a highly intelligent researcher might be able to remember the properties of the individual chemicals, attempting to calculate the effect of combining two or more chemicals rapidly multiplies the complexity of the research. With 100 chemicals mixed in equal amounts, there are roughly 9.3 * 10^157 (or 93 with 157 zeroes behind it) possible combinations of those chemicals (where order matters and repetition can occur). If those chemicals are mixed in unequal amounts, which is more likely, the number of possible combinations is nearly unfathomable.
By the way, making medicine is far more involved than just pouring chemicals into a bottle. A simple mistake in the order of mixing chemicals, the room temperature while doing so, and a host of other factors could turn a panacea into a poison.
A human cannot possibly compute all of the potential outcomes of those various combinations to produce the desired medicine in the first combination. Researchers are therefore obliged to make educated guesses about what might be a good combination of chemicals and then experiment. This is one of the reasons why pharmacology research can take years and millions of dollars to produce a potential solution or conclude that the desired medicine is impossible to create.
Artificial intelligence is a software tool that can accurately transform an intractable problem into a clear solution. We present Motorcycle™ as such a tool because of its ability to provide solutions to previously unanswerable questions via constraint satisfaction and breakthrough SAT solvability. We do not yet have the cure for the common cold, but, because of Motorcycle™ we can reduce the complexity of that search with our business partners.
We have created the world's most powerful deterministic polynomial-time AI engine.