I expressed these thoughts in a reply conversation responding to one of Kent Beck’s recent substack articles, very lightly edited here.

I understand the general desire to understand LLMs and to find out what they are good at and what they are not. And I think I understand your personal interest in them as well. I read your reports with interest and wish for “hard information” as to how good, how fast, and so on. I’m even tempted to try them myself, since all I’d have to do is click a button.

However. I have concerns:

Resource use.
This is reportedly high during use and immense during creation of the LLM’s knowledge, um base;
Personal cost.
I believe that most users will not look hard at the code their demon creates for them, and will thereby have less understanding of the program than they might;
Code quality.
It seems to me that an LLM cannot really do much better than average in its output, by the nature of their construction: this may not matter much: the margins on software are immense no matter how bad it is;
Human cost.
The LLMs are owned by the ultra-rich and their clear intended purpose is to replace people with AI wherever they can. In a world with universal basic income, and a general desire to take care of everyone, not just the rich, that might be a good thing. In this world, in my view, it is emphatically not a good thing.

One response to the above, even if one believes it, might well be to feather one’s own nest as best one can, learning to use and guide the “AI”, so as to preserve one’s position as long as possible. I can see that and understand it. It’s not what I’d call morally great, but it’s not horribly corrupt either, in my book. When faced with no good choices, we sensibly choose the least bad.

I am fortunate to be retired and no longer dependent on software to eat. And my best wishes go to those who are, and my honest acceptance of the choices they make in this new situation. I think LLMs are not a good thing for us and we need to find ways to live with them as well as we can.

So, bless you, bless us all, and good luck!