Comment on "proactive quantum secret sharing"

In the paper [Quantum Inf. Process. 14, 4237-4244 (2015)], Qin and Dai proposed a proactive quantum secret sharing scheme. We study the security of the proposed scheme and find that it is not secure. In the distribution phase of the proposed scheme, two dishonest participants may collaborate to eavesdrop the secret of the dealer without introducing any error. Incidentally, a possible improvement of the distribution phase is given.