T-Rex: Sometimes you read a story, and it's a cautionary tale! The whole point of the story is to tell you "don't do this, because here's a story where someone did that, and it worked out real bad for them."

T-Rex: And yet, we keep making mistakes!

T-Rex: And that suggests we're not producing SPECIFIC enough cautionary tales! Instead of "The Boy Who Cried Wolf" (moral: don't lie, because people won't believe you when you tell the truth, or alternative moral: tell better lies), we should have produced something more specific, like "The Boy Who Created A Social Network That Was So Broken And So Bad For Humanity As A Whole That It Incited A Genocide In Myanmar But That Same Boy Didn't Care To Fix It".

Utahraptor: That story could have saved us a lot of trouble!
T-Rex: RIGHT??

T-Rex: But with computers, we could have that! We could create an algorithm to expand general cautionary tales into thousands if not millions of SPECIFIC cautionary tales, thereby warning us of every possible threat that we might overlook if they were genericized!
Utahraptor: Let's do it!!

Narrator: LATER, T-REX'S ALGORITHM, TRAINED ON INTERNET DATA, BEGINS PRODUCING INSANELY RACIST CAUTIONARY TALES:
T-Rex: I recognize the irony in wishing I'd been warned about this danger!!
T-Rex: And yet, that doesn't stop me from wishing I'd been warned about this specific danger??