Google recently introduced its new "AI Overviews" tool, which is supposed to help reveal a quick answer to any search query without a user having to dig through search results themselves. It's a useful trick and one that works well in theory. The only problem is it's dead wrong all the time. 

It was reported early this year that Google would pay Reddit $60 million for access to its data to help train its A.I. models, and it seems those models are now pulling content from the jokesters of Reddit. 

Despite the tech world's instance that society-altering A.I. is right around the corner, they continue to demonstrate that even the most advanced systems have a ways to go when it comes to basic logic and reasoning. For example, this one seems to have a hard time distinguishing between animals, athletes, and sports teams with animal logos. Basketball is a particular weak point. 

Feeding the search bar incorrect information, or searching for something with popular Onion articles or satirical Reddit threads covering it, also seems to consistently produce misinformation. 

Maybe someday A.I. will run the world. But until then, we can laugh at its hilarious and consistent mess-ups.