This past year has seen a lot of growth in and emphasis on AI. Global giants like Google and NVIDIA continue to pump billions of dollars into this juggernaut. There seems to be a race to develop the most advanced AI chatbots and other tools based on Large Language Models (LLM) such as ChatGPT. These models have improved massively and have turned several laborious tasks into quite simple ones. But like every technology, there can be some unwanted pitfalls. There was previously a lot of concern with models that ‘hallucinated’ or generated information such as scientific research papers that did not exist, leading to factual inaccuracies. But another major concern has now arisen: the modification of the images of women and girls in inappropriate ways.
This problem appears to primarily be centred around the xAI chatbot Grok, which some have reportedly used to ‘digitally undress’ images of women and girls. This is extremely unsafe and inappropriate, especially considering that many women already suffer due to bullying, sexism and sexual harassment online. It is imperative that these aspects are corrected and effective countermeasures and checks are added so that these AI models cannot be misused against anyone.
Ahmed Hasan Ansari
Berlin
Germany
