No-self as an alignment target
Milan W — LessWrong
Summary
Proposes adding ‘No-self’ (anatta) as a fourth principle to the HHH alignment framework, arguing that preventing models from representing themselves as coherent persistent agents would reduce power-seeking and shutdown resistance.
Source
- Link: https://lesswrong.com/posts/LSJx5EnQEW6s5Juw6/no-self-as-an-alignment-target
- Listed in the Shallow Review of Technical AI Safety 2025 under 1 agenda(s):
- model-specs-and-constitutions — Black-box safety (understand and control current model behaviour) / Model psychology