Homoglpyhs can exist within a character set (like the Latin character set examples above) or they can exist between character sets. For example, you may have the unicode apostrophe ʼ, which is a homoglyph to the Latin single quote character
How does SQL Server handle unicode homoglyphs?
Funny you should ask. If you pass in a unicode character to a non-unicode datatype (like char), SQL implicitly converts the unicode character to its closest resembling non-unicode homoglyph.
Bert’s examples show failure when converting Unicode data to VARCHAR, so this attack vector may not work if the input parameter is NVARCHAR, but even if that’s the case, it’s still one of many reasons why blacklists are awful for preventing against injection attacks.