Hoes, E., Gilardi, F. (2025). Existential risk narratives about AI do not distract from its immediate harms. Proceedings of the National Academy of Sciences. https://doi.org/10.1073/pnas.2419055122