Those two features are present for a long in Bitcoin Core, but has been scarcely used over the years.
However, this is going to change with AOPP who directly link at this code in the protocol’s implementation.
Bitcoin Core will have better signing of messages once #24058 is merged
I don’t think there is anything wrong with this feature. If one or more exchanges use it differently users can avoid using those exchanges.
Signing messages has lot of usecases and sometimes even helps in verification without KYC. I have shared some examples in #10542 (comment)
NACK. Given the lack of SegWit / Taproot support in this feature, we should not assume the lack of usage is due to a lack of legitimate* interest. Also we can’t measure if it really is used “scarcely”; we could only infer that, indirectly, if for example nobody uses P2PKH addresses anymore.
That said, we should probably never implement aopp
URI support, at least not any VASP
specific features of it (and calling external URL’s is also problematic).
*
= for example to verify that a hardware device actually has the private key after you imported an xpub / descriptor.