public comms on Large Language Models (AI) like ChatGPT is mainly via journalists - and most of them hate it. But then I think their job doesn't really require a lot of what is great about LLMs & it does require what LLMs are bad at (fact checking).
LLMs are more of a game changer than google was!
LLMs are more of a game changer than google was!
Comments
Ask ChatGPT to list five articles by any Guardian journalist on some random subject you know they won't have written about.
You might be surprised by the results.
I think most people don’t really use them correctly.
I find they are incredibly useful for parsing lots of documents, e.g. using various long PDFs to fill out a form, or helping me refine my thinking around a problem.
For coding or "how to do" in tech they are invaluable.
LLMs are producing a lost generation.
"Yeah, you should fact check. Don't use it that way." True. But we are talking about vulnerable kids from backgrounds of only little education.
They are outsourcing critical thinking. In full.
Best case scenario is only two or three classes of students lost, until schools find a way to deal with it. Some countries are already returning to e.g. paper full time.
And in many regards dystopia is already the present.
https://medium.com/@alan-jones/uk-in-shock-ai-revelation-d22b42908179?sk=6532fe052b203cd9b20048a97b9c5f13
I think, like all AI, if you are an expert in something, the output is shallow and often wrong. If you aren’t an expert…
But if I just went with what AI wants me to produce, enshittifaction of software will get worse.