This touches two of my biggest concerns with artificial intelligence. A flood of AI generated fake content, and people outsourcing their thinking to AI tools.
I mean, if it wasn’t their AI tools they were outsourcing thinking to, it was the “smartest” person in their social circle. The question is if AI is smarter than that person for the average social circle
I recently visited an Australian dog breed info site. Most of the information there looked alright and very detailed. Pictures, history of the dog breed, physical traits, personality, uses, random tidbits, … It all looked well put together and reliable.
It’s a Pakistan hairless dog, that and the picture are about the only correct thing on there.
The history is basically the entire storyline of Starcraft, which might or might not have Khala dogs. Further down in the text it is also referred to as a horse, and a South-American bird of prey. Both of which I have no clue where it got that from, I did a very extensive search on South-American birds of prey and no Khala to be found. And it randomly switches back to dogs in-between.
So now I can’t trust anything on that site without it being verified through other means. Found more minor mistakes too after doublechecking other breeds, that I would’ve accepted without questions if I hadn’t stumbled across this gem.
So yeah, I find using AI for science/information purposes without doublechecking it by actual, human experts quite scary. This is just a dog breed site without much consequences, but I’m sure similar things happen (and slip through the mazes) with actual news too.
Outsourcing in more ways, than what another person will do for you. I think the Google Maps example tells exactly what’s going to come true when AI lets you easily skip any task that requires thinking, such as writing your homework.
This touches two of my biggest concerns with artificial intelligence. A flood of AI generated fake content, and people outsourcing their thinking to AI tools.
I mean, if it wasn’t their AI tools they were outsourcing thinking to, it was the “smartest” person in their social circle. The question is if AI is smarter than that person for the average social circle
But yeah, fake info is terrifying
I recently visited an Australian dog breed info site. Most of the information there looked alright and very detailed. Pictures, history of the dog breed, physical traits, personality, uses, random tidbits, … It all looked well put together and reliable.
Until I accidentally stumbled on the Khala dog page. You can read it here
It’s a Pakistan hairless dog, that and the picture are about the only correct thing on there.
The history is basically the entire storyline of Starcraft, which might or might not have Khala dogs. Further down in the text it is also referred to as a horse, and a South-American bird of prey. Both of which I have no clue where it got that from, I did a very extensive search on South-American birds of prey and no Khala to be found. And it randomly switches back to dogs in-between.
So now I can’t trust anything on that site without it being verified through other means. Found more minor mistakes too after doublechecking other breeds, that I would’ve accepted without questions if I hadn’t stumbled across this gem.
So yeah, I find using AI for science/information purposes without doublechecking it by actual, human experts quite scary. This is just a dog breed site without much consequences, but I’m sure similar things happen (and slip through the mazes) with actual news too.
That’s… That’s quite hilarious, to be honest.
Well, that’s terrifying. Learned some more StarCraft lore, though
Outsourcing in more ways, than what another person will do for you. I think the Google Maps example tells exactly what’s going to come true when AI lets you easily skip any task that requires thinking, such as writing your homework.