Their field is pretty much selling sloppiness-as-a-service, tho.
I'm genuinely a bit concerned that LLM true believers are beginning to, at some level, adopt the attitude that correctness _simply does not matter_, not only in the output that spews from their robot gods, but _in general_.
It's kinda crazy to witness, you can see in the main GPT-5 release thread that there are people excusing things like the bot being blatantly wrong about Bernoulli's Principle in regards to airplane flight. I wish I could find it again but it's thousands of comments, one of the comments is literally "It doesn't matter that it's wrong, it's still impressive". Keep in mind we're discussing a situation where a student asks the AI about how planes fly! It's literally teaching people a disproven myth!
I'm genuinely a bit concerned that LLM true believers are beginning to, at some level, adopt the attitude that correctness _simply does not matter_, not only in the output that spews from their robot gods, but _in general_.