Not everyone is going to agree what the best defense is going to be but regardless of what the puzzle opponent does, your first move is always going to be the same.
It's possible the puzzle is going for a specific pattern. It could be that the defense is the only one that delays mate for the longest time. It's also possible the same position is in the database with multiple different defenses
I am not sure which forum category this should be in, sorry, and maybe this topic has been discussed a hundred times already. But has anyone noticed that sometimes in the puzzles the opponent's response to your move is not really the best move they could make? So when trying to solve the puzzle you might reject a possible move because the opponent has a good response to it, only to find out the "solution" is in fact to make that move after all and then the computer makes the opponent play a worse response. Below I give an example of this (with spoilers) in an actual puzzle on chess.com.
This really ruins the puzzle by making it virtually impossible to find the "solution", because you have no way of knowing that the opponent will make a less than optimal move. In some cases it could mean there isn't really a valid solution at all because you apparently can't achieve your objective under best play.
As an example, look at puzzle 1026480:
Spoilers below:After 1. fxg3 fxg3 2. Nf3 the "correct" solution is 2. ...Rxf3. The computer then responds 3. Qe3 and the puzzle proceeds from there culminating in a series of checks. But this is not White's best move (as verified by Stockfish)! If they play 3. gxf3 instead, then it is not at all obvious what Black should do next, and although Black still has the advantage it is not clear how they recover the loss of their rook.