The article explores the vulnerabilities of detecting AI-generated media and the use of watermarks to protect digital content. It discusses the limitations of visible watermarks, such as compromising the art and discouraging audience use, as well as the potential misuse of stock photos with watermarks. Imperceptible watermarks are then introduced as a solution to protecting intellectual property without compromising quality, and addressing the issue of AI-generated content flooding the internet. However, the article highlights that imperceptible watermarks have vulnerabilities, as they can be removed using AI algorithms. The process of removing both perceptible and imperceptible watermarks using AI is explained, and researchers are highlighted as developing countermeasures to known attack algorithms. The article also suggests that legal consequences may be applicable when removing watermarks from AI-generated images. Ultimately, the article concludes that while watermarks might not stop bad actors, they could be beneficial for good actors, and highlights the need for further research into developing more robust watermarking systems.