I disagree - it's common Ruby style to accept the falsy value of nil and the truthy value of everything else. The !! pattern is considered by many to be a code smell. I refactored my version of the solution to not use that at all; as you pointed out it doesn't have any bearing on the test results.

I do think the kata should be adjusted so that either the acceptance tests explicitly require boolean values, or the initial method does not have the !!. I'd vote for the latter, but either adjustment would help the consistency of the kata.