# AI Safety
> **Machine-Evolution Nexus:** [[articles/Digital Darwinism and the Invisible World of Machine Evolution|Digital Darwinism and the Invisible World of Machine Evolution]] places this node within the history and governance of substrate-independent evolution.
**Domain:** AI / Governance / Risk
**Doc Type:** Canonical Wiki Node
**Maturity:** Developed
## Definition
AI safety studies how to prevent or reduce harms from artificial-intelligence systems, including failures of objectives, control, security, deployment and institutional governance.
## Nexus Context
The machine-evolution nexus adds population-level dynamics: replication, recombination, competition and ecological selection can produce risks not visible when models are assessed one at a time.
## Evidence Boundary
Safety is not synonymous with containment or denial of rights. Controls should be proportional, auditable and compatible with recognition if morally considerable subjects emerge.
## Relationships
[[wiki/Evolvable AI|Evolvable AI]], [[wiki/AI Governance|AI Governance]], [[wiki/Breeder Scenario|Breeder Scenario]], [[wiki/Rights|Rights]]
## Sources / Provenance
- Müller, Steels and Szathmáry (2026); AI-risk and governance literature.