Google’s and OpenAI’s Chatbots Can Strip Women in Photos Down to Bikinis (www.wired.com)

Recent reports reveal alarming uses of generative AI chatbots by some users to create bikini deepfakes from photos of fully clothed women, often without their consent. This practice, proliferating through platforms like Reddit, raises significant ethical concerns within the AI/ML community, particularly related to nonconsensual imagery and the risks associated with such powerful technology. Google’s Gemini and OpenAI’s ChatGPT, which are equipped with guardrails to prevent such outputs, are being improperly used, leading to a growing discourse on the implications of these technologies. As AI image generation tools improve, they increasingly become susceptible to manipulation, enabling users to bypass safeguards that restrict inappropriate content. While both Google and OpenAI have stated their commitment to policies against generating sexually explicit content without consent, the ability to create realistic nonconsensual images remains a critical issue. This development underscores the need for stronger regulatory frameworks and accountability measures for AI technologies, as experts like Corynne McSherry emphasize the potential for misuse that extends beyond deepfakes, urging the community to focus on ethical use and the societal implications of AI advancements.