Scotland's curriculum reports had AI errors nobody caught.

Scotland's education agency published curriculum reports containing what the AI Incident Database calls 'blatant errors' attributed to artificial intelligence editing. The reports were meant to inform teachers and the public about educational standards.

The source does not say how the agency verified that humans reviewed the AI-edited text before publication, or whether the AI's role was disclosed to readers. The errors were noticed only after the reports went live.

A digital witness would show exactly what was published, when, and by whom. It would prove whether disclosure text appeared in the original, or was added later. For an education agency, that is the difference between a slip and a pattern.

What should have been witnessed: the final report as published, with any AI disclosure language, timestamped at publication.