[james_a_craig]

Come on, you need to encode it if you're counting bytes not characters, even if that's equivalent for this particular string. :)

8*len("I'm wrong".encode('utf-8'))