I have given a one-hour lecture on AI technology in English for international students.

Naturally, this is the first story I discussed at the beginning of this lecture.

This is what I have been saying and will continue to say.

That aside.

Before the lecture, I had a short talk with the students.

The contents are as follows.

(1) They usually don't do programming.

(2) When they use AI technology, they use packaged software.

(3) They only did "Pascal" in a lecture in college.

Frankly, I was quite shocked to hear the story.

"Researchers who do research using AI technology do not code AI technology on their own"

This is the same situation as in the country I belong to (Japan).

And now that I know that the same situation exists outside of Japan, I honestly can't believe it!

("Pascal"? That's a 30 year old paradigm. At the very least, they should be taught Python.)

-----

'How are they supposed to understand AI technology without tracing and debugging the programs in the first place?' I could not hide my shock.

I remembered having such a conversation with a junior colleague once -- "For them, the output of packaged software is the final solution, and they don't think beyond that(*)," he said.

(*) I could not find the page in my column If you find it, please let me know.

-----

Programming education in our country's compulsory education process is simply a waste of time."

This "apprehension" of mine has now risen to the level of "certainty".