AI Governance Club
🌍 AI Governance Club
Most AI risk discussions focus on rogue systems or sudden takeovers. This paper makes the case for a different scenario: one where humanity loses control gradually, through thousands of individually reasonable decisions, as AI displaces human labor and cognition from the systems that govern our lives.
Join the AI Governance Club for a Special Author Presentation and live Q&A with one of the paper's co-authors:
Gradual Disempowerment: Systemic Existential Risks from Incremental AI Development (2025)
Kulveit, J. et al
The paper argues that as AI takes over roles in economies, cultures, and states, the mechanisms that kept those systems aligned with human interests quietly erode, without any single catastrophic event. This session will explore what that process looks like, why alignment of individual AI systems is not enough, and what governance interventions could make a difference.
Session format: Author Presentation & Q&A with the author
Presenter: David Krueger
Professor David Krueger is the founder of Evitable and a leading voice on the societal risks posed by advanced AI systems. He was a founding Research Director at the UK AI Security Institute, initiated the landmark CAIS Statement on AI Risk, and is a co-author of the influential "Gradual Disempowerment" framework, which explores how humanity might gradually lose control of AI. After more than a decade at the forefront of AI research, he now focuses on halting the pursuit of superintelligence by mobilizing public awareness around these risks and calling for a moratorium on AI development.
For more information and to register for upcoming sessions: AI Governance Club