It feels like every progression in computer vision is improving so drastically! The quality of Imagic's results is very convincing. I don't need to mention the high potential for abuse of this ability, because the authors acknowledge this as an inherent problem in their Societal Impact section (thank you to them for doing this!): "However, as with other approaches that use generative models for image editing, such techniques might be used by malicious parties for synthesizing fake imagery to mislead viewers. To mitigate this, further research on identification of synthetically edited or generated content is needed." I agree with the sentiment that identification and required disclosure of AI-altered images are very important to regulate, especially as this technology and many others will become more widespread.