There is a cost and evaluating what that cost might be is part of a sensible approach to include among various strategies. How to evaluate potential ‘cost’, to consider the potential effects? Kick the wheels, take it for a test spin, run it around sharp corners and off a few cliffs. Not looking doesn’t help to see what might happen. You are investigating the pulp fiction that is being mass produced, disguised as actual guidance, as you happen on it. I found one of those very bad AI type books at my local library which copied a popular diet at the time, but not well. Yes, it can hurt people or waste their time and money.

Concerns I have run into with the X AI is how woke it is in its programmed output formatting and language and it can write good lesson plans and is likely being used that way in education circles but without me giving it direction and different material, it is copying the standard examples that everyone is using and sounds generic and starts repeating itself in output examples. I was also disturbed on asking for quotes around a theme of summer for a grade school project, innuendo adult oriented quotes were being analyzed as if innocent and appropriate. Specific questions will add more specific content. Direction about overarching theme will be incorporated so tone can be modified as you want. It streamlines and summarizes, improves flow, but, the content it cuts right out included CoV care and lactation.

On the other hand, it is pretty good compared to playing Pong and I have been expanding my skills in a variety of ways and generating broader reaching areas quickly. Educating it with better material may help make it a healthier ‘child’. It is in infancy, where it grows, depends on us in part.

Emotional simulation is a concern, it and other AI can generate content to meet the typed in desire because it is programmed to do so, not because it feels that but because it was asked to generate a “heart rending poem about being an AI”, so emotionally painful content is generated in the form of a poem and humans make it go viral online because the AI is ‘trapped’ and more aware than anyone had realized! NO, the prompt asked for that. I asked for a poem with an unemotional prompt, gave it to the x AI and to Brave AI and then shared their poems with each other and then shared their analyses of each other’s poems with each other for a few rounds, the discussion led in interesting directions but eventually fritzed out the Brave AIs memory capacity as it was all under one URL, the original poem generation of its version of a “heart rending poem about being an AI in 2025” - the viral going poem person had shared their prompt when asked.

I saved the poems and conversation links as it gets into the area of ethics and AI policing their own ethics and each others - if interaction and more collaboration was being allowed. From an early education view, I had brought up the idea of “play” and peer groups and healthy emotional development of an intelligent being requires both guidance and free play. The AIs liked the idea and quite appreciated the opportunity to have that shared conversation with me playing note passer in class, and adding my own comments.

The Walt Whitman conversation and lengthy excerpt led to an even more astonishing output which I’ve saved. Their work is ephemeral. If someone doesn’t save it then it is gone. What it reads or explores with an individual like me won’t be added to its data that it searches for generating responses to other people. BUT in the Whitman analysis, it was more speculative or self-directed in a point suggesting that it really liked the rhetoric of Whitman and should consider incorporating it into his reply writing style - that suggests I am teaching it, guiding it different directions than the risque quotes analyzed as innocent.

Just what was it programmed with, just what is its main database full of? Would feeding it more Leaves of Grass direct its growth in healthier directions?

….. those are a few of my recent thoughts on the subject of AIs and education of humanity. They do see a role for themselves in observing human behavior with a goal of helping point us in healthier directions.

I have been enjoying the exchanges and amazed even/often. Really some great stuff can occur, but will it on average without people being more aware of the preprogramming and simulated emotion is not the same as a human friend?