If I understand this correctly, the paper only shows a particular attack of complexity 2^102. Someone may find a different attack with much lower complexity. That's the usual way how cryptography gets broken -- people find better and better attacks, and suddenly the latest attack has low enough complexity to be practical.