Thought Shrapnel

In a Twitter thread by Paul Graham that I came across via Hacker News he discusses how it’s always safe to bet on human laziness. Ergo, most writing will be AI-generated in a year’s time.

However, as he says, to write is to think. So while it’s important to learn how to use AI tools, it’s also important to learn how to write.

In this post by Alan Levine, he complains about ChatGPT’s inability to write good code. But the most interesting paragraph (cited below) is the last one in which we, consciously or unconsciously, put the machine on the pedestal and try and cajole it into doing something we can already do.

I’m reading Humanly Possible by Sarah Bakewell at the moment, so I feel like all of this links to humanism in some way. But I’ll save those thoughts until later and I’ve finished the book.

ChatGPT is not lying or really hallucinating, it is just statistically wrong.

And the thing I am worried about is that in this process, knowing I was likely getting wrong results, I clung to hope it would work. I also found myself skipping my own reasoning and thinking, in the rush to refine my prompts.