herbAIrium

A 4-minute AI-generated video self-portrait. Both the voiceover and the character are AI-generated, with the voice trained on the artist’s own voice and the character modelled from selfies. This work examines the normative portrayal of human beings generated through AI video technology, highlighting the biases inherent in the datasets used to train these AI models. These biases reflect limitations in image databases, raising critical questions about how AI mirrors or exacerbates societal prejudices in generating human representations.

The work also makes a broader point: the issue of bias is not exclusive to AI. Analogue systems, such as the botanical classification methods of traditional botanists, have historically imposed their own biases, often resulting in a lack of diversity. This emphasises that classification and representation challenges transcend AI technology and have long existed in human knowledge systems.