Maybe it was the categorical nature of "no danger whatsoever" that led to the comparisons to religion. Given the difficulty of predicting anyone's psychological development, and given that you yourself say that you've seen multiple lapses before, what rational reason could you have for such complete confidence? Of course, it's true that there are things besides religion that cause people to make predictions with probability 1 (which, you must concede, is a plausible reading of "no danger whatsoever"). But, in human affairs, with our present state of knowledge, can such predictions ever be entirely reasonable?