he's not wrong, it is stupid. that stupidity rises from the possibility of you maliciously altering your rule. according to your logic, #38 is also a simple rule. how do we know you didn't just change your mind? that's not how we tackle problems empirically, you see, we aren't afraid that some kind of a higher being is going to change how something works, because we solved it once. Now maybe that is the case, but why should we concern ourselves with that? if we gave that possibility too much weight, we couldn't progress, because we would be afraid of everything being unreliable, dynamic. Basically, since the answer to your solution is based on randomness and completely unregulated, you have the complete power to decide whether someone is correct or not and that contradicts with nature of mathematics quite a bit, considering there is no element of trust in there.
i get that you are emphasizing how several values taken form a list that follow a certain pattern don't guarantee it, basically the problem with empirical evidence, but i believe the task you presented doesn't represent that very well