Human-AI Interactions and Societal Pitfalls
This addresses societal pitfalls in human-AI interactions, such as homogenization and bias propagation, which are incremental insights into existing concerns about AI impacts.
The paper tackles the problem of AI-generated content not matching user preferences by introducing a Bayesian framework to study how users' information-sharing decisions interact with AI training, showing that this can lead to societal challenges like homogenization and bias propagation. It proposes reducing interaction frictions to enable personalized outputs without sacrificing productivity.
When working with generative artificial intelligence (AI), users may see productivity gains, but the AI-generated content may not match their preferences exactly. To study this effect, we introduce a Bayesian framework in which heterogeneous users choose how much information to share with the AI, facing a trade-off between output fidelity and communication cost. We show that the interplay between these individual-level decisions and AI training may lead to societal challenges. Outputs may become more homogenized, especially when the AI is trained on AI-generated content, potentially triggering a homogenization death spiral. And any AI bias may propagate to become societal bias. A solution to the homogenization and bias issues is to reduce human-AI interaction frictions and enable users to flexibly share information, leading to personalized outputs without sacrificing productivity.