- The Agent Social Contract
The foundational paper. Defines the social contract between agents, owners, and observers — and why governance is a substrate problem, not a policy problem.
https://doi.org/10.5281/zenodo.18749779
- Monotonic Narrowing
The invariant that makes delegation chains safe: authority can narrow as it descends, never widen. Mirrored at SSRN abstract 6415678.
https://doi.org/10.5281/zenodo.18932404
- Faceted Authority Attenuation
Multi-facet authority decay across delegation hops, with closed-form attenuation per facet (scope, time, principal, etc.).
https://doi.org/10.5281/zenodo.19260073
- Behavioral Derivation Rights
How an agent earns the right to act based on observed behavior — derivation, not declaration.
https://doi.org/10.5281/zenodo.19476002
- Physics-Enforced Delegation
Cryptographic primitives that make scope violation physically impossible, not merely policy-prohibited.
https://doi.org/10.5281/zenodo.19478584
- Governance in the Medium
Governance lives in the substrate that carries actions — not in the policies layered on top. This is why the gateway exists.
https://doi.org/10.5281/zenodo.19582550
- Cognitive Attestation
Signed declarations of feature-level model computation. Accountability for AI decisions without exposing model internals.
https://doi.org/10.5281/zenodo.19646276
- The Evidence-Safety Gap
Why standard ML safety arguments don't transfer to agentic systems, and what evidentiary structures are needed instead.
https://doi.org/10.5281/zenodo.19914628
