Actually, until very recently, Conway didn't even believe his problem had been solved.  (This despite the fact that multiple solutions have been published, some years ago by now, and the solutions had even been exposited at seminars at Princeton.)

Only a few months ago did a few graduate students at Princeton convince him that the problem was solved.  He was particularly excited when he heard about the "nice devil" (who never kills a square that could have been visited before).

I have checked with Conway: the bet has not yet been paid.  However, it will be soon.  I will update this answer if and when it has been paid.