Yes, that is what they're referring to. The technical details behind it are kind of cool. It's a "bug" of the JBIG2 compression algorithm. See 1, 2

[1] https://community.apryse.com/t/jbig2-compression-issue/1814

[2] https://en.wikipedia.org/wiki/JBIG2#Character_substitution_e...

There‘s also a great conference talk [0] on this, with an english translation [1].

[0]: https://youtu.be/7FeqF1-Z1g0

[1]: https://youtu.be/zXXmhxbQ-hk

Yes, several national archives where affected on all settings. If a institution scanned and archived documents with these- these documents are - in theory challenge-able in court- within reason..