I don't like guys who give orders without knowing the hardships in the field (well, who doesn't like them).

I, too, am on the side of such "giving orders."

However, it isn't easy to realize that you are standing on the side of giving orders.

People are indifferent to the fact that they are "the ones giving orders" and are resented and hated by their subordinates and others.

We are unaware of the pressure we put on our people, making them sick and retiring them.

On the other hand, humans feel uncomfortable when they are being "ordered around." And we never forget our hatred for that human being.

(Sometimes, in TV interviews, those who summarize with beautiful stories such as "In hindsight, my boss's reprimand" are "good-natured personalities" or "self-protective liars" or "falsifiers of their history").

Whenever human beings are organized (socialized), they become like this.

And we know that there is hopelessly no way to fix this.

In this society, it is tough to escape from "the one who orders" and "the one who gives orders."

-----

Recently, however, I have discovered that there are means by which I can become aware that I am on the "commanding side."

Generative AI.

Recently, I have been using "ChatGPT" or "Github Copilot" for most of my programs.

I rely so much on generative AI that my coding skills may decline.

(I mean, I think it's declining.)

Here is how I currently make my program.

======

(Step.1) Input the outline of the program in Japanese to the generating AI.

(Step.2) Execute the generated program

(Step.3) If an error appears, enter it into the generating AI and return to (Step.2).

(Step.4) If no error appears but I cannot get the expected result, enter that into the generating AI and return to (Step. 2).

======

I would repeat this and continue until the program works as expected.

However, no matter how often repeated, the program may not work.

Then I want to ask the generative AI, 'Why isn't it working? -- even though I'm getting help without having read the program thoroughly.

This attitude is the same as the behavior of the people I despise the most.

Thus, I became aware of the "ordering side".

-----

No matter how harsh I am, the generated AI will not become "depressed" or "retire" or "commit suicide" -- this, I think, a sense of security will probably make humans 'arrogant.'

I am concerned about the harm caused by generative AI in a different direction than many people think -- losing jobs to AI.

Generative AI will likely turn ordinary people into "aggressive" and "impulsive" people.

The logic that it is OK for a "generative AI" to be "aggressive" and "impulsive" because it is a tool that complains and obeys without complaint,

It is equal to the logic that it is OK for a "feeble-minded person" to be "aggressive" and "impulsive" because they are human beings who submit without complaint. It is the same logic that justifies "bullying" and "power harassment.

In other words, it is the same logic that justifies "bullying" and "power harassment.

-----

Therefore, I recommend the following mechanisms in the generative AI:

"If a human being cannot rationally handle a generative AI, the generative AI will refuse to provide services to that human being."

It is not "for the sake of generative AI" but "for the sake of us humans."