As much as we hate to admit it, a lot of us get caught doom scrolling from time to time.

And just so you know, it's not your fault.

Those little devices are designed to suck us into their bottomless rabbit holes.

They want you to click, to buy, to engage, to comment, and of course, to like.

It's good for business.

But too much of it isn't good for our mental health.

Some carefully curated feeds are meant to make you feel not so great about yourself,

to think you can be prettier, smarter, healthier, even richer.

I know I don't need to tell you all this, but that's not reality. However, our brains are

easily tricked into thinking it is. What's terrifying is, we're not the only ones who are being

exposed to misleading, incorrect, or

harmful information online. AI is also learning about the world by processing information,

reading news articles, and analyzing photos. It's a lot more human than we'd like to admit.

And if you give it bad information, inaccurate facts,

or fill it with psychologically disturbing ideas, that shapes the way it sees the world.

Which brings us to an artificial intelligence called Norman. He was created in 2018 and trained on a lot of data that was

considered unsettling. When Norman's programmers ran it through a few psychological tests,

they proved it had symptoms of psychopathy, which shows AIs are capable of psychological depth.

They are capable of wants, desires, needs,

and even show signs of mental illnesses. And if that's the case, who's to say they won't

decide to pursue their goals in the future? Maybe at humanity's expense. I'm Yvette Gentile.

And I'm her sister, Rasha Pecorero. Welcome to another episode of So Supernatural.

...