Homogenization or correlated failures in model derivatives
AI systems that fail to perform reliably or effectively under varying conditions, exposing them to errors and failures that can have significant consequences, especially in critical applications or areas that require moral reasoning.
"Homogenization refers to common methodologies and models used across down- stream GPAI systems, which may lead to uniform failures and amplification of biases [176, 30]. This risk arises when numerous downstream AI systems are built upon a few large-scale foundation models."(p. 51)
Supporting Evidence (1)
"This may be caused by centralization of AI advancements within a few compa- nies, as well as flaws from algorithmic monoculture, where dataset sources and collection methods are similar across numerous AI models. Homogenization in models may lead to consistent and arbitrary rejection, mis- treatment, scrutiny, or misclassification of specific users of groups, as well as the spread of implicit perspectives (e.g. bias towards a particular political group) across multiple application domains."(p. 51)
Part of Impacts of AI (Bias)
Other risks from Gipiškis et al. (2024) (144)
Direct Harm Domains (content safety harms)
1.2 Exposure to toxic contentDirect Harm Domains (content safety harms) > Violence and extremism
1.2 Exposure to toxic contentDirect Harm Domains (content safety harms) > Hate and toxicity
1.2 Exposure to toxic contentDirect Harm Domains (content safety harms) > Sexual content
1.2 Exposure to toxic contentDirect Harm Domains (content safety harms) > Child harm
1.2 Exposure to toxic contentDirect Harm Domains (content safety harms) > Self-harm
1.2 Exposure to toxic content