PESAEAS

Policy-Enforced Secure Adversarial Enclave for Autonomous Systems

Peh-SAY-uhs

A global initiative for autonomous system containment.

About

PESAEAS advances policy-enforced adversarial enclaves — hardened environments that assume the contained system may try to escape, deceive, or exploit its boundaries, from software agents today to physical systems tomorrow.

The initiative convenes technical discussion through meetups and shared practice, toward better containment standards over time.

For systems engineers, security practitioners, AI safety researchers, and anyone building isolation for autonomous software or robots.

An initiative by Akioud AI, Paris. Technical work is intended to be open as the community grows.

Related open-source work includes Akios and EnforceCore.

Focus

Adversarial threat model

Assume the contained agent may try to escape, deceive, or exploit its boundaries.

Policy-enforced boundaries

Safety as declarative, auditable rules — not ad-hoc checks or hope.

Enclave isolation

Hardened containment from software agents today to physical systems tomorrow.