I've warned several times over the years that these links will rot and our going to be out of our control, but... that's what they want!
Right now there are around 540 TLDs that anyone can use (not brands or country restrictions) and most are really cheap and haven't been bought out yet.
Be careful if you care about visits from those using corporate networks, many arbitrarily block uncommon TLDs unless users explicitly ask for the restriction to be dropped (and even then an assessment is made, the drop is not automatic and your site alone is unlikely to be deemed sufficient reason).
Unlike link shorteners, the URL length barely matters, and the codes often get permanently printed onto posters, tables, art, etc.
the QR they offer (free today, then you need to pay) is a total scam: but so many professionals cannot spot it until is too late (saw in printed magazines, books ...)
Have a look at trustpilot reviews
Consider https://www.somebusiness.example/multiple-long/parent-direct.... This will be a version 9 code (53×53) with medium error correction. (You could also make it version 8 (49×49) with low error correction, but that’s riskier.)
You might think I’m exaggerating with that URL; I wish I was.
By contrast, HTTPS://QR.EXAMPLE/XYZABC can be the smallest and clearest size, version 1 (21×21) with low error correction (which is normally acceptable at that size).
The short URL will scan more easily and reliably than the long one.
(I’m mildly sad that HTTPS://CHRISMORGAN.INFO/ already completely fills the 25 characters available when you use the alphanumeric set, so my home page is the only version 1 QR code I can use.)
short.name. 300 IN SHORTN 0 5 /foo https://some.long.name/whatever/foo
...could tell a browser/app/whatever that http(s)://short.name/foo is supposed to point to https://some.long.name/whatever/fooAt the very least, we could have a standard shortener API call that clients could use to verify before allowing the link redirect.