Send message to Author

Paper short abstract

Roko’s Basilisk posits a future AI that punishes those who didn't create it. I analyze this as a dark mirror to Silicon Valley utopianism. Instead of liberation, it offers entrapment in deterministic logic, revealing how digital communities ritualize the surrender of human agency to the Algorithm.

Paper long abstract

This paper investigates “Roko’s Basilisk” – the internet thought experiment positing a future superintelligence that retroactively punishes those who failed to facilitate its creation. While often dismissed as “tech-bro folklore,” I argue the Basilisk functions as a potent vernacular theology that reveals how digital communities are actively reconfiguring the boundaries of moral obligation across time. Drawing on the notion of implicit religion in AI narratives and anthropological critiques of longtermism, I analyze how the Basilisk narrative creates a temporal feedback loop. In this loop, a hypothetical future intelligence is granted immediate ontological weight, stripping current human subjects of agency and reducing them to “standing reserve” for a machine god. Through a discourse analysis of online rationalist communities and reaction threads, I explore how the “horror” of the Basilisk is not merely a fear of punishment, but an epistemic crisis. The paper demonstrates that the Basilisk serves as a dark mirror to Silicon Valley’s utopianism: instead of liberation from the body, it offers an entrapment in a deterministic logic where human worth is measured solely by one’s utility to the Algorithm. Ultimately, I contend that studying such information hazards is crucial for an anthropology of AI, as they expose the fragile affective architectures supporting our definitions of sentience, causality, and the human itself.