The AI Image Generator: The Limits of the Algorithm and Human Biases
2022 has been the year of AI image generators. Over the past few years, these machine learning systems have been tweaked and refined, undergoing multiple iterations to find their present popularity with the everyday internet user. These image generators—DALL-E and Midjourney arguably the most prominent—generate imagery from a variety of text prompts, for instance allowing people to create conceptual renditions of architectures of the future, present, and past. But as we exist in a digital landscape filled with human biases—navigating these image generators requires careful reflection.
Midjourney AI image generated using the prompt "African Architecture". Image © Midjourney AI / Matthew Maganga
2022 has been the year of AI image generators. Over the past few years, these machine learning systems have been tweaked and refined, undergoing multiple iterations to find their present popularity with the everyday internet user. These image generators—DALL-E and Midjourney arguably the most prominent—generate imagery from a variety of text prompts, for instance allowing people to create conceptual renditions of architectures of the future, present, and past. But as we exist in a digital landscape filled with human biases—navigating these image generators requires careful reflection.
