Idk about how impressed to be with uh… parsing the image. Like there is a bottle that says soy sauce, so it read that and then just…. Other ingredients? Vary specific!
It is not impressive, but it makes the rest of the output even worse. You’re expected to treat the bot’s output as human language, but it doesn’t make sense like language would: it identifies the soy sauce, it should be able to identify the bowl is empty, no change happened, and yet it’s still babbling that the guy “already combined the base ingredients”.
Idk about how impressed to be with uh… parsing the image. Like there is a bottle that says soy sauce, so it read that and then just…. Other ingredients? Vary specific!
It is not impressive, but it makes the rest of the output even worse. You’re expected to treat the bot’s output as human language, but it doesn’t make sense like language would: it identifies the soy sauce, it should be able to identify the bowl is empty, no change happened, and yet it’s still babbling that the guy “already combined the base ingredients”.
Look, it’s just excited to grate the pear, okay?
:3
Is that a euphemism?