Hey Felix, picture a chessboard where every policy about AI is a piece on the board. I see a pawn moving too early on e4, risking a fork that could turn into a blunder in the endgame. How do you feel about timing when we roll out new AI tech?