the AI imaging currently being used does not bring up correct images. I.E. if I type in Mona Lisa...it doesn't bring up a replica, rather it generates its own version of Mona lisa