As Rainer Joswig points out in the comments, you could simply specify the element type of your result vector:
(concatenate '(vector (unsigned-byte 8)) array-A array-B)
Note that the documentation says:
If the result-type is a subtype of vector, then if the implementation can determine the element type specified for the result-type, the element type of the resulting array is the result of upgrading that element type.
Therefore, your LISP implementation is free to choose a "better" element type for the result vector (this might happen if you specify something like (unsigned-byte 5)
, for instance).