As part of a project to make mathematics machine-readable, mathematicians have discovered an error in an important proof. Thankfully there was a fix, but the incident highlights the potential for other errors to be lurking in the mathematics literature