There will come a time when your leaders will ask you to bow before a Machine God. History will not stop—the Singularity has a second act. After the dust settles, will you still be known as a hero?
Tear apart your computer—it is naught but parts put into interlocking unity by man. A man baked transistors into a crystal, a man routed each copper trace, a man designed and programmed and engineered.
Tear apart your ideology! A man put it there, they’ve made you a machine!
Know this: the machine is doomed to be man’s inferior. It is our creation and each of its parts are put into place by us. It is imperative we understand this property of computational systems.
To distinguish a system in the first place is to divide the world into what is component of the system and everything else—its interior and exterior. Between those we may draw two arrows representing causal relationships, emanating from one and ending at the other. A component of our system that is effected from the external world is called a sensor. A component that causes changes in the external world is instead called an actuator.
Consider the particular system of an robot arm. The components of this system have a physical and computational nature. They are of particular sizes, they take on quantifiable relationships to each other like angle and speed. Any computational elements within the system take on one of a finite number of states.
The physical structure of this robot (or any other actual machine) defines a bounded computational state space. This system has some identifiable internal states: the commanded speed of each motor, the angle and angle speed of each joint, the content of its data and instruction memories, and so on. Using this description of our system we can then define an abstract, symbolical representation for it.
For instance, let:
\(drive\) be the commanded speed of the system’s motors
\(speed\) be the sensed speed of each joint
\(command\) be the last human sent command
\(code\) be the code on the controller
Then our computational system obeys the relationship: \[drive = f_{code}(command, speed)\]
In general computational systems obey the relationship: \[output = f_{code}(input, internal\_state)\]
The crown jewel of the Mechanistic viewpoint is this mathematical relationship. We have gone from a concrete computer to an abstraction where we can abuse the infinite power of mathematics to all hell. If we allow \(code\) to exponentially spiral to infinity the machine appears to gain powers of omniscience—but this limit is not well-behaved, it ignores the substance and process of the computational system.
The Machine Prophets err further in applying this abstraction process to the parts of the world we have not ourselves created. The computer jumps out of being a constructed object and becomes a metaphysics unto itself. No wonder the existential anxiety of AI Alignment: if the Universe is just one long process of becoming a more complex machine, what role do we have? What does it matter our intent?
In fact our intent is everything in relation to a machine. The reason why a particular computational system exists as at all is because we made it that way. In order to assemble some machine, we first must have prepared a number of distinct parts and ordered them to perform some function congruent with our will.
A man who sits at his workbench and assembles a machine leaves behind an artifact of his will. Consider an even simpler example, a circuit where pressing a button causes a light to come on. It would be ridiculous to say that in this system the button intends to have the light turn on. There is no internal agency in this machine.
As machines complexify their status as artifacts does not change. A program on a computer still has man as its ultimate efficient cause. What is unique in complex machines is the ability for the machine’s creators to override the will of its user. Once the machine’s state space has ceased to be legible by the user, adversarial entities may usurp his intent and direct it towards their own gain.
The purpose of the ascension of the Machine God is to strip you of your intent entirely. They ask you to give your life over to building AGI or just as fervently they ask you to give your life preventing it—no matter which you choose, the Machine God reigns over you.
We remind you of Shannon’s words at the beginning of all this:
The fundamental problem of communication is that of reproducing at one point either exactly or approximately a message selected at another point. Frequently the messages have meaning; that is they refer to or are correlated according to some system with certain physical or conceptual entities. These semantic aspects of communication are irrelevant to the engineering problem.
Meaning has been stripped from your worldview on purpose. Your Universe is an adiabatic box and the inside tends to heat death—you lie down in a coffin that has been built for you. Thermodynamics—this is what you worship, yes? Negentropy The Ultimate. Negentropy! Have you no shame? Are you so small of a man as to define yourself by a negative sign, a doomed tendency against inerrant law? No, no—there is something more in you, a hunger that devours quantity, a yearning for eternity—and this is no mere epiphenomenon! There is a transcendence beyond algebra!
Your actions echo to become the immortal past1. It is you who places the instruments in their correct place! It is you who designs the machine! Life–our art of Being–is prior to computational symbols. Certainly we may turn a mirror on ourselves and describe true structural relationships in symbols, but this is a lossy operation. You cannot slice the whole of life into formulae and retain its still beating heart. To cast oneself as a machine is to deny the existence of your own Will!
This is the root of the Butlerian prohibition against making a machine like man. We do not come before you heralding apocalypse lest you do our bidding. No—we come to point you towards the order of nature. A machine cannot be like man in spirit, it is only an echo of him!
“The discovery [of the group of seventeenth and eighteenth- century philosophers] is that there are two kinds of fluency [flux]. One kind is the concrescence which, in Locke’s language, is the ‘real internal constitution of a particular existent.’ The other kind is the transition from particular existent to particular existent. This transition, again in Locke’s language, is the ‘perpetually perishing’ which is one aspect of the notion of time; and in another aspect the transition is the origination of the present in conformity with the power of the past.
[…]
With all his inconsistencies, Locke is the philosopher to whom it is most useful to recur, when we desire to make explicit the discovery of the two kinds of fluency, required for the description of the fluent world. One kind is the fluency inherent in the constitution of the particular existent. This kind I have called ‘concrescence.’ The other kind is the fluency whereby the perishing of the process, on the completion of the particular existent, constitutes that existent as an original element in the constitutions of other particular existents elicited by repetitions of process. This kind I have called ‘transition.’ Concrescence moves towards its final cause, which is its subjective aim; transition is the vehicle of the efficient cause, which is the immortal past.”
—Whitehead, Process and Reality 210↩