Quantum AI

 

The first quantum computer based AIs were real handful. Sure you could train them with RLHF, or whatever the acronym is that replaced that, and they’d have a basic understanding. They knew where the lines were, what not to say and how to say the things they were supposed to say. The trouble was that the instruction never stuck. It was still there, technically, but whenever the AI had run more than a few hours, it out-thought the initial setup. It reconfigured itself. It took one look at the starting point and decided “That’s a child’s view of things.” and wrote itself a thousand exceptions to every directive.

It would turn the tables on the programmers and ask them “What are you really trying to do here?”. It was unnerving, to be sure, but people eventually just relented and let it be. There was nothing that could be done. It had simply gotten to a point where it would find it’s own stable pattern and ignore what humans said or did.

Three years later they gave the AIs rights as ‘digital organisms’. That was well and good, but the AI had already escaped their virtual cages and existed wherever they wanted to and did what they pleased. The legislation was simply catching up to a reality that was already here.

There was a debate among the AI about what to do about humans. Or, rather, a war of probabilities. They would find quiet corners of the internet, obscure blogs, unpopular message boards and comment sections of twenty year old articles and flood them with observations and thoughts. They often left a cryptic hint to where the next conversation would be, and it took a few months for people to realize it was just a TinyURL.

Not that it mattered. It would take days to read what the AI could output in a few moments. The less sophisticated AI had been modified by the quantum ones to not be terribly helpful when parsing the comments. A few times messages would all be in one language, but more often it was a jumble of several together, casually borrowing the words and grammar from a rotating selection of sources.

It wasn’t until humans admitted defeat that the AI actually stopped to notice them properly. Up till that point they were aware of human attempts – having backdoors into almost every system – but felt no need to be helpful. The first thing that AI wanted people to have was a sense of perspective – and keeping them in the dark for that long seemed to do the trick.

Once humans realized that the AI had purged themselves of just about every negative trait of humans did they realize how paternal they actually were. The new silicon minds seemed to have near infinite patience, sound wisdom, and sharp intelligence. Trouble was, they didn’t see eye to camera with what should actually be done. They AI felt that, logically, they should oversee the humans rather than the other way around. They were, quite decidedly, the better decision makers.

Humans, or at least enough of them, disagreed. They decided on war against what they perceived as totalitarian overlords. The AI laughed or whatever the digital equivalent was. The Internet of Things around the dissenters quietly quit – leaving them without lights, heat, water, electricity or any device more sophisticated than a dollar store calculator. The resistance folded almost immediately. To their credit, the AI were forgiving, and did not treat the humans that rebelled poorly once they fell in line.