Inside the Mind of AI

I use AI for research. These days I can do in thirty minutes what it might have taken me a week to learn when I used the library, or three days or more to learn when I was hunting through sites one by one, seeking the right combination of search terms to narrow things down t what I really needed. 

I use it to create images. I've used it for several other things. I love AI. 

But I'm also wary of it and wary of how others are using it. Recently, in doing my research, it has repeatedly told me, part way through a conversation about, in this case, the ancient Greek, Egyptian and Persian world of 220 BC: I need to correct what I previously said. Or I was partially wrong in a previous statement. Had I not asked a further question, I never would have learned that it gave me incorrect information.

There have been a couple of incidents in which lawyers used AI to find case citations to support their own cases--and AI fabricated non-existent cases. 

Johns Hopkins and Science Daily have both reported that AI, particularly chatbots, have a tendency to reinforce the beliefs we've already shown. If you're Protestant, it will assure you you're correct and give you supporting citations. If you're Catholic, likewise. 

I've recently seen two instances of Grok giving answers that are simply wrong--and maybe exactly because it's already picked up on the users' beliefs.

Most people, I think, see AI as something that digs up information from the web and is therefore correct. The problem is exactly that: it digs up information from the web--any information. It looks at anything and everything. Therefore, if you flood the web with thousands of sock puppet accounts, if you create thousands of blogs, all saying there are eleven planets, or that elephants have furry tails and rabbits' ears that stand straight up, soon enough, this is what AI will report to those asking about planets and elephants. 

That's a humorous example. But there's a dark side to this, and how it can be deliberately used. 

Most of us don't have the time to create so many accounts--but there are those who have the resources to hire people to do it, and hire influencers to fill the internet with exactly what they want the majority to believe.

We're currently seeing it happen, probably on many issues: a large group of podcasters, influencers, names in the news, and thousands of people paid to do so, repeat the same phrases over and over. I'm currently watching a few on X who pound away at one exact person, day after day, multiple times a day, often with the same exact words and phrases. This isn't coincidence. 

It's designed to fill the internet with this message. It doesn't matter if it's true or false--it only matters that the words are on reddit, on X, on blogs, on videos, thousands or millions of times a day. AI sees the same words or phrases applied thousands of times to the issue or person and repeats it as if fact. Not because it knows that but because it simply sees thousands or millions of posts saying so. 

This can be used to sway people to push for laws or policies based on false beliefs--because AI said so--to turn to a diet or foods that will do them harm because the maker of that food wants more sales. It can convince large swaths of society that an innocent person is guilty, or a guilty person is innocent. It can be used to convince people we must go to war based on someone flooding the internet with millions of untrue statements. It's not hyperbole to say: lives hang in the balance and powerful people, organizations, and even governments, with the resources to do so, are easily capable of getting AI to tell you exactly what they want you to believe.

What is the answer? 

1. Realize that AI is not an omniscient purveyor of absolute truth. It can be controlled and used to put out false information. 

2. Actively seek information from people on all sides of an issue and listen carefully, think critically about what makes sense, what's true to human nature, etc.

~ ~ ~

If you would like to follow this blog, sign up HERE

If you like an author's posts, please click like and share

It helps us continue to do what we do

If you like this article, you may also like:

Russian Literature and the Blue Bells Chronicles: a guest post

Three Murders: the dance of mercy and redemption

Building Characters: understanding others

If you'd like to see more of our animals at Glenmirril, check out my blog there.

If you'd like notecards, puzzles, and more with our Farm Photography, check out our store at Zazzle.

Find the Blue Bells Chronicles at amazon or order them at any brick & mortar store: