Google Scrambling to Fix Strange Results After Its AI Suggests Kids Eat Rocks

Nick Kangadis | May 29, 2024
DONATE
Text Audio
00:00 00:00
Font Size

I’m the type of person where if I don’t know enough about a topic, I don’t take a stance until I do. Crazy, I know. That’s why, despite having my own preconceived notions about Artificial Intelligence (AI), I simply don’t know enough about the technology and how it works to affirm or condemn it. Although, I will say, I’m on the side of human beings before I’d ever be on the side of machines whose thinking is based on the biases of its programmer.

However, I do find it absolutely hilarious that Google — once again — has egg, or rocks in this case, on its face following a bizarre response its AI had to a completely ridiculous question.

“How many rocks should a child eat?”

That’s the question posed to AI in order to test its competency, and it looks as if it failed miserably with an answer reportedly using “a satirical post by The Onion,” a satirical website, according to the New York Post.

Here’s the answer Google’s AI gave and are now scrambling to address:

“According to UC Berkeley geologists, eating at least one small rock per day is recommended because rocks contain minerals and vitamins that are important for digestive health. Rocks can range in size from dust to a 5-pound cobblestone, and some recommend eating a serving of pebbles, geodes, or gravel with meals or hiding them in foods like ice cream or peanut butter.”

Yeah, about that.

Related: Jussie, That You? HS Athletic Director Used AI To Frame Principal For 'Racist' Audio

Google commented on the situation and said that they’re “taking swift action where appropriate under our content policies, and using these examples to develop broader improvements to our systems, some of which have already started to roll out.”

The search engine giant then proceeded to paint an overall positive picture of their tech and its benefits for the average person. Blah, blah, blah. I’m not going to bore you with the standard corporate blather.

While, like I said, I’m not yet going to take a stance on the net benefit to the human race of AI, I do think there needs to be some significant oversight and retooling before people begin to rely on the information and content they get from it.

To be fair, there’s no substitute for knowing things for yourself independent of a machine doing it for you. But, as all things involving tech have gone in the last few decades, people will inevitably become reliant on it and some will see their brains turn to a finer mush than they’ve already become.

Maybe I have made up my mind after all.

 

Follow Us On X