← Back to all Desirable Properties
DP12Active

Community-based AI Governance

AI systems in the Meta-Layer are governed not by corporations–but by the communities that use them.

9 Second Call alignments

2 extensions

1 clarifications

Overview

Meta-communities play an active role in governing AI systems, ensuring their actions align with community standards and ethical principles. This collaborative oversight promotes accountability and trust in AI-driven interactions and decisions.

Why It Matters

Meta-communities help shape the ethics, behavior, and evolution of AI. This isn't top-down control–it’s participatory oversight that keeps intelligence aligned with integrity.

Key Elements

AI Governance via Meta-Communities

Meta-communities can play a role in overseeing and governing AI behavior, ensuring that AI operates in ways that align with community standards and ethical practices.

Zone-Scoped Governance

Communities define rules within specific zones of interaction, matched to the context and risk of what happens there, with explicit boundaries and inheritance between nested zones. Governance applies where the interaction occurs rather than as a single global setting.

Policy as Executable Objects

Rules are expressed as structured, machine-readable objects carrying scope, triggers, enforcement hooks, authorship, and version history, and they bind to behavior at the moment of generation, moderation, ranking, or data access. Binding is deterministic and inspectable, so identical inputs under the same policy produce the same governed outcome.

Governance Loops and Memory

Governance runs as a continuous cycle of propose, implement, observe, contest, and revise, with stated time bounds and clear state transitions. Decisions, rationale, disputes, and outcomes are persistently recorded and linked to policy versions, so communities can learn from prior rulings instead of relitigating them.

Policy-Bound Verification and Containment

Governed actions emit receipts naming the policies applied, the conditions evaluated, the outcome, and any override, and those receipts are independently verifiable. Containment mechanisms enforce what governance declares through permission gating, quotas, sandboxing tiers, and graduated escalation, with drift between stated intent and observed behavior treated as a detectable defect.

Participatory Ratification with Bounded AI Assistance

Adoption is a defined event with stated quorum and thresholds, a signed policy object, recorded rationale, and a notice period proportional to the change's impact. AI may summarize, simulate, and analyze proposals, and must disclose that assistance, but material decisions require human ratification.

Workgroup

Creating community-driven governance models for AI systems that ensure transparency, accountability, and collective oversight.

Second Call for Input

Community submissions from the Second Meta-Layer Call for Input that aligned with, clarified, or extended this property. These are historical provenance–not live governance votes or comments.

9 alignments

2 extensions

1 clarifications

Aligned submissions

Clarifications

Extensions

Explore the on-chain Call for Input archive

Early PCI Conversations

People Centered Internet email threads (Sep–Nov 2023) that contributed to this property’s development. Each is permanently inscribed on Bitcoin via Ordinals.

1 aligned

View all inscribed PCI conversations →