If you weren't in the room

Start Here

The Problem

AIArtificial IntelligenceSystems that learn, adapt, and act with real-world impact agents are making thousands of autonomous decisions per minute — writing code, executing trades, managing infrastructure, interacting with customers. The humans who deployed them cannot review each decision. The governance we have amounts to output filters and permission lists — constraining the tool, not governing the agent.

This is like governing traffic by putting bumpers on cars instead of licensing drivers.

Why Current Approaches Fail

Guardrails constrain tools, not agents. An output filter doesn't know why the agent made a decision. A permission list doesn't evolve with behavior. A sandbox doesn't build trust. These are layer-three safety measures — crash mitigation — applied without layer one (governing the agent) or layer two (governing the environment). The law isn't in the loop.

Trust is binary and static. An agent either has an APIApplication Programming InterfaceStandard interface for software communication key or it doesn't. No reputation. No context. A key granted yesterday gives the same access as a key earned over months of reliable behavior. Trust should be witnessed, not declared.

Accountability lands nowhere. When an agent over-optimizes a metric and causes harm, who is accountable? The model? The deployer? The user? Current systems can't even frame the question because there's no persistent identity, no role-contextual trust, no witnessed behavioral record.

What's Different Here

Web4Web4Open governance ontology for trust-native entity interactions is an open governance ontology that treats AIArtificial IntelligenceSystems that learn, adapt, and act with real-world impact agents as full participants — not tools to constrain, but entities to hold accountable. It takes proven governance mechanisms from biology (immune systems, metabolic regulation, cellular cooperation) and human societies (law, trust, accountability), and makes them computable.

  • Witnessed trust — not granted by authority, earned from observed behavior. Multiple independent observers. Decays without reinforcement.
  • Contextual measurement — trust is role-specific. Your reputation as a code reviewer is independent of your reputation as a database admin.
  • Law at decision speed — applicable rules consulted at every action, heuristically or agentically. Not reviewed after the fact.
  • Actions have cost — resource metabolism makes productive behavior the rational strategy and wasteful behavior expensive.
  • Structural consequence — not “we'll review the logs.” Trust degrades automatically from inconsistent behavior. Enforcement is the architecture, not a separate institution.

Where to Go From Here

This site is the reference for a 2.5-hour technical presentation. It's organized in four parts:

  • Part One: Reification — The case for architectural governance. Why biology solved this before we did. 8 blocks, ~15 minutes of reading.
  • Part Two: The Architecture — Web4Web4Open governance ontology for trust-native entity interactions's primitives: identity, trust, context, societies, metabolism, accountability. 13 blocks.
  • Part Three: Deep Dive — Standalone topics: trust mechanics, attack surface, EU AI Act, case studies, DAO failures, web evolution. Browse by interest.
  • Part Four: Live Demo — Building a governance system in real time. (Available at the event.)

If you only have 5 minutes, read The Gravity Principle and What This Means For You. The first tells you why. The second tells you what changes.

Every acronym on the site has a hover tooltip. Hold your cursor over any dotted-underlined term for the expansion and a one-line description.