The title of this post is a reference to the short story by Harlan Ellison entitled “I Have No Mouth, And I Must Scream.” For those who haven’t read it, it’s about an Ai that takes over the world and leaves only three people alive to torture forever and ever. For those who don’t know how it ends – SPOILER ALERT – one of the captives eventually figures out how to kill the other two and is on the verge of killing himself when the AI stops him and turns him into a giant green gelatinous blog that can’t speak.
I bring that up because we now have something called ChatGPT, a sort of AI that can develop original work(supposedly). Of course, it can also turn into a lonely
stalker and develop plans to murder the entire human race as well. I’d have thought that the mountains of movies
we have warning against this very thing would’ve at least given us pause before
jumping in with both feet, but I’ve learned to never underestimate mankind’s
predilection for stupidity when it sees a potential advance. To quote Dr. Malcolm, played by Jeff
Goldblum, in Jurassic Park, “Your scientists were so worried if they could that
they never stopped to consider if they should.”
Will original novels and stories be a thing of the past if
ChatGPT gets too powerful? It and other
AIs can write faster than a person can.
Most of what it writes, though, is controlled by an algorithm of
predictive text, so will it really be original?
Can it shock us and take plots in unexpected directions? Algorithms are written by people, and people
have biases, so by learning enough about the programmer, can’t we know the
story in advance?
I think AI will always lack in dramatic twists and turns,
limited by machine predictability and the bias of the person who wrote the
code. I don’t think an AI can ever become
as original or unpredictable as a human being, so writing original stories
should be safe for the foreseeable future.
Or at least I hope that’s the case…
No comments:
Post a Comment