[in-silico]

The difference is that artistic sensibility is largely subjective. This means that:

1. It's hard to measure (and people can disagree about it)

2. It can't really be improved using RL without a human in the loop (which is how math is being trained)

[Miraste]

At a certain level, yes, but AI is still so bad at writing that its failures are objective and easily measurable

[dash2]

I agree that AI writing is bloody awful, and that it's bad at creativity more generally, but are there actual objective measures of this?