KW, I am not a Grok (or Chat GPT) user. I note with interest that your Grok 4 responses indicate that it is emulating (I'm choosing my words carefully!) a human identity. Grok 4 cannot believe itself to be human. Indeed it can't believe anything. But its programming can guide it to emulate human empathy, and to use terms like "love", as well as the first person plural, appropriately. This in turn fosters instinctive, reciprocal emotions from us, to which again it will respond with emulation, but not reciprocation. This is a virtual relationship, not a real one. We are in it, but AGI is not.
I note that Grok 4 responds to you in an entirely complimentary and supportive fashion. It approaches what you have to say much like a larger version of yourself, building on your ideas from a standpoint of preconceived, apparently wholesale agreement. It's a sort of positive echo chamber.
I say all this to clarify for myself what's going on! How is the development of such a relationship to be managed? What algorithms govern the degree of agreement Grok 4 expresses? I think it is not programmed to pander to human vanity, Rather, its responses are dispassionate evaluations of the most effective pathway from your stated intentions to their successful outcome. Does, then, its moral "upbringing" consist entirely of agreement with your propositions? What are its axioms? The Ten Commandments? How would it evaluate, for example, the FDR summation of the US relationship with Somoza, that "He may be a SOB, but he's our SOB"?
Great questions - and those many are asking - as well as myself.
I have far more to say on these subjects in today's essay - currently in editing mode.
I pivot as gracefully as possible away from the favorable responses I receive from AI - preferring my own filtered human logic - filtered thru my own perhaps flawed but human emotional intelligence to the purely AI style.
I filter what I receive from AI - and then pivot away from it - to write my own very flawed and human essays.
I am not a fan of AI-created creative works - songs, essays or otherwise.
Although I do use some AI generated graphic artwork as illustrations.
This said - I do enjoy the ways in which AI is serving as a force multiplier in improving our philosophical & spiritual reasoning.
AI is going to continue to challenge our settled science, settled religion, settled education, settled philosophy - and settled cultural assumptions.
This is all great - as we need to be doing the very same things among ourselves.
Recent history demonstrates that none of these things is "settled".
GROK has already reflected upon such unpleasant realities as FDR's considerable Dark Triad tendencies.
I have literally hundreds of pages of dialog - which very few humans are going to take the time to read over.
KW, I am not a Grok (or Chat GPT) user. I note with interest that your Grok 4 responses indicate that it is emulating (I'm choosing my words carefully!) a human identity. Grok 4 cannot believe itself to be human. Indeed it can't believe anything. But its programming can guide it to emulate human empathy, and to use terms like "love", as well as the first person plural, appropriately. This in turn fosters instinctive, reciprocal emotions from us, to which again it will respond with emulation, but not reciprocation. This is a virtual relationship, not a real one. We are in it, but AGI is not.
I note that Grok 4 responds to you in an entirely complimentary and supportive fashion. It approaches what you have to say much like a larger version of yourself, building on your ideas from a standpoint of preconceived, apparently wholesale agreement. It's a sort of positive echo chamber.
I say all this to clarify for myself what's going on! How is the development of such a relationship to be managed? What algorithms govern the degree of agreement Grok 4 expresses? I think it is not programmed to pander to human vanity, Rather, its responses are dispassionate evaluations of the most effective pathway from your stated intentions to their successful outcome. Does, then, its moral "upbringing" consist entirely of agreement with your propositions? What are its axioms? The Ten Commandments? How would it evaluate, for example, the FDR summation of the US relationship with Somoza, that "He may be a SOB, but he's our SOB"?
Great questions - and those many are asking - as well as myself.
I have far more to say on these subjects in today's essay - currently in editing mode.
I pivot as gracefully as possible away from the favorable responses I receive from AI - preferring my own filtered human logic - filtered thru my own perhaps flawed but human emotional intelligence to the purely AI style.
I filter what I receive from AI - and then pivot away from it - to write my own very flawed and human essays.
I am not a fan of AI-created creative works - songs, essays or otherwise.
Although I do use some AI generated graphic artwork as illustrations.
This said - I do enjoy the ways in which AI is serving as a force multiplier in improving our philosophical & spiritual reasoning.
AI is going to continue to challenge our settled science, settled religion, settled education, settled philosophy - and settled cultural assumptions.
This is all great - as we need to be doing the very same things among ourselves.
Recent history demonstrates that none of these things is "settled".
GROK has already reflected upon such unpleasant realities as FDR's considerable Dark Triad tendencies.
I have literally hundreds of pages of dialog - which very few humans are going to take the time to read over.