Take AISafety.info’s 3 minute survey to help inform our strategy and priorities

Take the survey
Beyond the basics

Language models
Mesa-optimizers and subagents
Decision theory
Mathematics of agents
Strategy and outcomes
Brain emulation
Human intelligence enhancement
Computer science
Values
AI consciousness

What is a singleton?

A singleton is a world order in which there is only one decision maker that controls most significant outcomes, as opposed to a multipolar scenario

where there are multiple relevant actors. This decision maker would be able to prevent any other actor from threatening its continued control and would be able to achieve its desired outcomes unopposed. The decision maker may have internal complexity and be composed of multiple subagents: it could be a single dominant AI, a group of collaborating AIs, a small group of humans with AI assistance, a nation, or even “a strong set of global norms that include effective provisions for their own enforcement”. A singleton could be robustly aligned to the will of a person or group, or pursuing objectives that are not aligned with any human.

There are numerous pathways which could lead to a singleton. One example is if there is an extremely fast takeoff

and the first superintelligence self-improves until it can overcome all opposing forces.

The concept was introduced by Nick Bostrom

in 2005.

Keep Reading

Continue with the next entry in "Beyond the basics"
What is a multipolar scenario?
Next


AISafety.info

We’re a global team of specialists and volunteers from various backgrounds who want to ensure that the effects of future AI are beneficial rather than catastrophic.

© AISafety.info, 2022—2025

Aisafety.info is an Ashgro Inc Project. Ashgro Inc (EIN: 88-4232889) is a 501(c)(3) Public Charity incorporated in Delaware.