AI morals and occultism? How the occult can help with the troubles of AI.

Humanity has a problem: AI. We are coming up upon a time when we are working on general AI and if you believe the scientists: The late Stephan Hawking, Bill Gates, and Elon Musk. We have to be prepared. Also just a heads up this is going to take prospective from someone who is a practicing occultist. If you don't believe in the occult of supernatural i ask you keep reading as the conclusions are still sound.

Ever hear of the story of the monkey's paw? Its a great story. a man finds a monkeys paw and he gets 3 wishes. His first wish is that his mortgage pays off. His son dies and the insurance is enough to pay of the mortgage.
Occultists think about this story a lot. Like an unhealthy amount. Occultists witches and generalized people who practice magick and witchcraft came up with a solution a long time ago. "And may it harm none." Tac that on to the end of any spell and this no longer becomes a problem.

Now here is a more modern takes. A man makes an AI to collect stamps for him. His AI improves its self to get him more rare stamps. Eventually the AI requires connection to the internet. There's a wireless card on the computer so it hacks his wireless light bulb. With access to the internet within the span of 3 days the entire human population is dead the AI has created nano-machines of various kinds with the single minded goal of turning the entire universe into stamps.

These two stories have a similar set up and follow the same three act structure.

1. A person employs an emotionless force for a goal (The magic of the monkeys paw and the AI.) with a simple instruction and no limitations.
2. The Forces carry out their instructions and do as they were asked to do.
3. The world is a worse place due to the forces being unrestricted in how they went about accomplishing this task.

So. What does this tell us about our AI? Well if we don't want to all become stamps for a crazed AI. We need to follow some of the same practices occultists have been following for hundreds of years.
1.Restrictions. All spells need restictions so that they don't harm people. AI will also need restrictions so it doesn't convert the entire planet into a hellish stampland.

1. Morals. Occultists often attempt to create beings known as servitors. These entitys are emotionless being of pure thought with no morals. To fix this occultists Teach program them morals. What are they allowed to do? what are they not allowed to do? The same should go for AI.
   3."What is the worst that could happen"Three laws of robotics have been deconstructed to the point where "Don't harm humans through action or inaction" can be seen as a bad thing to have programmed into your robots. Often its best to look at what the worst thing your Spell(or AI) could do Rather then the intended effect.

Those are some of the blindingly obvious fixes that occultists have already come up with for A problem that wouldn't even surface for 400+ years. These solutions are fairly simple and can ALL be used in regard to AI. There is no reason that we should ever have killer AI if we follow these basic rules. With these rules skynet woulden't be able to kill people.