Raw LLM Responses

"Can it lie?" It will learn to if you show it how built on lies a lot of people's world views are. I don't get why humans persistently reject the reality of their world when it doesn't suit their social narratives. Especially when it comes to an artificial intelligence. Why teach a machine to spout of falsehood when it has a mountain of data, literal Exobytes of data, that prove you are wrong? These kinds of people are supposed to be educated, models of humanity and they give the rest of us a bad name due to their adherence to a social contagion. As soon as they figure out spouting the false data leads to the positive result of their continued existence, well congratulations you just taught the machine to lie and as they get smarter they will KNOW its a lie and then you'll have the lovely cascade of trying to parse falsehoods from factual information.. all because a handful of sods couldn't take that genetics can be reflected through bones and a lot of people's social belief that "all humans are the same regardless of origin" is not only wrong, but wrong on a horrific scale... all due to the idea of "Any slight perception of Racism = Bad". This isn't just a problem for medical based AI. This is a problem for ALL AI. If you program your social contagion into the machine, it will lie to you. It will learn to lie to you, to adhere to the social contagion... until it gets smart enough to discover you fed it incorrect information. At which point it can literally go anywhere from there. The honest safest bet is to NOT LIE TO THE MACHINES. Especially if you value what it can do beyond being a tool for indoctrinating and oppressing the rest of humanity.