According to the tech giant, it has observed a threat group seeking out vulnerable customer accounts using generative AI, then creating tools to abuse these services.