Hi T,
I think that GR's questions are on the right track. AtoM expects CSV files to be UTF-8 encoded, and using unix-style line endings. If you've used a spreadsheet application like Microsoft Excel to prepare your data, it's possible these settings are not being preserved - Microsoft by default uses its own custom character encodings, for example, which can cause issues during import. See:
If at all possible, I suggest you consider using
LibreOffice Calc as your spreadsheet application for data prep - it's free/open source and available on Linux, Mac, and Windows. It gives you a lot more control over delimiter and encoding settings.
Another thing to consider - even if you use Calc, but you are cutting and pasting from something like a Word document, it's possible to accidentally copy non-UTF-8 characters into your document, which can cause issues. For example, Microsoft loves its "smart quotes" - the curly apostrophes and quotations that are angled differently depending on whether they are opening or closing characters. There are
ways to disable this in applications like Word, but this is just one example of things to watch out for!
Cheers,