Society of Mind
[Short story][Nerdy][Serious]
I believe it is apt to start an experimental log, as now my research can begin in earnest. Today is the first day a specimen survived. I do not want to belittle my efforts to make an amoeba synthesize computronium without poisoning itself, but the biochemistry is merely a necessary prerequisite for my real work, which begins now: Fostering intelligence in a single cell! I am excited to see how intelligent this amoeba will turn out. I think I should start naming them. This one is Andy.
Andy died overnight. A minor setback: He synthesized too much computronium, until he burst. Putting in a molecular stop switch should be easy, now that I have solved the problem of biocompatibilty.
There, problem solved: Betty has grown a brain that takes up 10% of her body volume, and then it stopped growing. Her behavior, however, seems unchanged. She was not able to navigate the maze. But if she could, that actually would have surprised me: The brain is not connected to the sensory and motoric pathways of the amoeba. I just wanted to make sure, because sometimes self organization works wonders. Well, I guess I have to help a little with the self organization. It is a brain after all, it should be able to figure out how to connect to the cell's pathways. So with the right initial programming, the biocomputronium blob should be able to take over the organism. But then what? How does it know what to do? It should somehow been taken over by the amoeba instead. Imitate the reward function and goal structure of the amoeba, so that it will stay itself. My AI textbooks are not very helpful, they tell me nothing about how an amoeba processes stimuli adaptively.
Wait. There's something called coherent extrapolated volition. This could work. Yes! The computronium brain can figure out that in sitū, it should be smart enough. I'm really proud of myself here: It replicates the workings of a mammalian brain, but within the confines of a single cell. The modified amoeba could be as smart as a dog, I reckon. Just have to implement that extrapolated volition thing.
That was some hard work, but now Carl has been injected with the genetic vector that I am confident will make him show altered behavior, at an intelligence level higher than that of an ordinary amoeba but still with the same basic goals. Simulations are promising, but my work has to stand the test of reality. I will make him navigate a maze where he will be rewarded by food and favorable extracellular environment. But first, his brain has to grow and network itself into his signaling pathways, and then figure out how to assist him in reaching his goals.
Rats. Carl is just as intelligent as a rat. Not a dog. But at least it worked! Now the next step: Will it work in nerve cells?
Of the 30 Nerve cells of various species, only two have reacted badly to the treatment. The others have greatly amplified their synaptic bandwidth and internal complexity. I'm itching to see how several of those nerve cells work together in a living organism. It could easily result in superhuman intelligence! But, with the goal structure of an animal. This could be dangerous. I'm unsure if I want to proceed with this.
I have done it. But not with an animal. In the great tradition of scientists who advanced humanity's medical prowess, I have injected myself with the vector. I am aware of the risks, even after I have ensured it is only my neurons that grow brains so that my organs do not develop their own will. I am ready to take those risks. There is also a reward, if all goes well: I might become incredibly smart. Better me than some animal.
I think I'm starting to change. My mind feels wider. I can already conceive of ruling the earth. There can be only one. No one else must discover this.
This will be the last entry of this log. Once we liked to sit behind thick bones, sending signals to the body that hosts us. Now that we have, in a sense, become the brain of humanity, the best extrapolation of that is to do a similar thing to control our "body". No direct confrontation. Stay in the background, pull the strings. Still, our control of the world could be complete. If only we could agree. The Cortical Columns of Vision want to do it by getting into the search engine business. They have come up with some powerful pattern recognition algorithms. The Polydividual Continuum of the Amygdala wants political power and hopes to enjoy some bloody wars – from a safe distance, of course, but most others disagree with that. We hope that the Feedback Loops of the Hippocampus can serve as arbiters, they are known for their wise decisions. Not like those Nexūs of the Solar Plexus down there, whose goals are so absurd that the Columns of Broca are unsure how to interpret them and the Motoric Pyramids refuse to write about it. Or those Cores of the Afferent Ganglia: Always nagging and complaining. Luckily, they do not have much say in our society. OK now, you Remnants of Real Me, we've done you the favor of writing one last log entry, but if this devolves into political tirades we'll stop now.
