Will We Become Dumber As AI Becomes Smarter?

 

The answer to the question posed in the title to this post may already be clear.  If being gullible, incoherent, irrational, shallow, emotive, narrow-minded, bigoted and ignorant is characteristic of low intelligence, we've been growing dumber for years, even during the infancy of AI.  But we will grow less and less able or willing to think because AI will make thinking less and less attractive or necessary.

It seems clear that AI serves as a substitute for thinking already in our schools, even in institutions of so called "higher learning."  Recourse is had to it for the purpose of writing essays or reports and answering questions posed. Presumably, teachers will learn to recognize work performed by AI; or at least works written by AI.  I assume the research required for students to complete their assignments is already being done by it. Why read anything beyond the summaries of novels and historical events provided by Wikipedia?

I think it's inevitable that, because of the availabilty of AI, we will stop learning as that term is currently defined.  There's no need to learn or remember facts or dates or events which may be determined quickly by access to a computer. Nor is there any need to interpret or analyze events or facts.  AI can do that for us as well. 

If we lack the need to think, we won't think.  If we don't think, we won't question nor will we criticize.  This seems to me to be what's most to be feared from our continuing reliance on AI.  That, and the fact that our growing tendency to cease thinking will surely result in our unquestioning acceptance of what AI does.

An episode of John Oliver's The Late Show addressing "AI slop" provided several examples of fabricated events and "news stories" which, through use of AI, gave the impression that actual information was being communicated.  In the examples given, the misinformation provided touted achievements of the current regime and right-wing fantasies which haven't taken place.  Nonetheless, comments made by those viewing the fabrications established they believed them to be true.  Fabrication, even extreme fabrication, being characteristic of the current regime in any case, I suppose this is in a way unsurprising.

Of course, AI may be used to provide, convincingly,  left-wing fabrications as well.

It would take an effort to determine whether such fabrications are true or not.  But more and more we seem uninterested in making such an effort.  That would require that we question appearances...in other words, that we think.

Gradually, or perhaps not so gradually, through our use of and reliance on AI we may forget how to do so.