mistake corrected

An inconsistency made in Cairo in 1924 got remedied in Madina.
The typesetter in Būlāq made a mistake in 2:264.
ʿUṯmān Ṭaha copied it faith­fully.
It was first corrected in the UT pc set version of the State of Kuwait, printed in Khartum,
later in UT3 of the King Fahd Complex.
The hamza should not sit on the tooth.

Oh — actually the other way round.

here as most of the time, the maġrīb makes it as it should be.

I want to look at the "problem" in a general way.

In 4:112 we have with /ī/, while in /riʾāʾ/ there is just a short /i/: after /ī/ the hamza sign hovers over the connection; when there is no /ī/, it sits on the tooth.

Above I mentioned that Otto Pretzl pointed out the inconsistency in writing /riʾāʾ/, but his point was that ad-Dānī has the hamza one the tooth.
And he was concerned by another group of mistakes. ad-Dānī writes the dual without alif, the KFE mostly with alif.

the first line is BnF arabe 328a

riʾāʾa ‒ sometimes like this; sometimes like that

Already Otto Pretzl had noticed that in Gizeh1924 riʾāʾa was not written the same at the three places it occurs ‒ without the authorities pre­scrib­ing that.
Iranians and UT2-KFC have the first hamza always after the tooth.
Muṣḥaf al-Azhar aš-Šarīf (Cairo 1978‒1986?) has it always on the tooth:

The others have it sometimes on, sometimes after the tooth.
In 2:264 it is normally on the tooth, except UT2-KFC and Iran.
In 4:38 it is normally after the tooth, except Taj-KFC ‒ probably changed in the next edition (1440h. it is still on the tooth).
In 8:47 all have it after the tooth except Muṣḥaf al-Azhar aš-Šarīf.