³¹It is obviously important to consider the exact definition of free will. In one radical viewpoint, freedom of will is a matter of not being physically or psychologically forced or compelled to do what one does. This viewpoint is called “compatibilism” in philosophy since it implies that free will is compatible with determinism (Strawson and Watson, 1998). Consider a basic robot. If it decides to raise its arm, is there any physical constraint that would prevent it from doing so? Are the computations only due to its own sensory input (in its cameras or else) and computations made in the processors inside the robot? Here, the idea of free will is formulated in terms of what causes the agent’s actions: Is it solely information-processing inside the agent—in a modern formulation—or is something outside it influencing the decision? If so, even such a robot could be said to have free will. Humans would certainly have free will. It may not be the conscious self that decides actions, but the neural networks in the brain. Still, as long as the neural networks are inside the human skull, it is, the human that decides and controls its actions. Yet, many find such a definition of free will questionable. These include all the schools in the philosophy of free will other than the compatibilists. What I described in the main text is rather similar to the “pessimist” school. Namely, an obvious counterargument to the compatibilist definition is that is depends on the time scale used: we should look back in time, trying to find the original reasons for your actions. As I argue in the main text, fundamentally, the robot’s or human’s actions are just a result of its programming/evolution and, especially, the input data, so there goes free will.