For some time I've been meaning to read back through my blog, at least a little bit, to try to process and synthesize it and latch hold to some of the most salient points and themes. Of course I'll never read the whole thing. That would be ridiculous.
So here you have these AI large language models and you would think they might be good for a task like this. Read the blog. Think about it in relationship to other analogous bodies of work, say perhaps Samuel Pepys's diaries or Dooce's blog. I will confess that part of me was loath to open my kimono wide open to the LLMs, particularly after Sarah Silverman and others are filing suits against their creators for theft of intellectual property. Then again, hers is worth money.
ChatGPT wouldn't even approach the task. Google's Bard at first was fired up, even going so far as to say "I am excited to learn more about your blog and to have a dialogue with you about it" and asked some specific questions. But it really wasn't able to do much, quickly crawling back into its shell by saying "I am just a large language model and don't have the capacity to help with that."
I'm sure I could go back and be more concrete with it and coax some thoughts out of it, but I'm not sure that's the highest and best use of my time. Maybe. For now, it's time to pack it in and go have dinner with Mary.
Here's the original and undoubtedly most generative AI.
No comments:
Post a Comment