About Kybernology

A new field of inquiry emerging at the intersection of AI systems, cultural studies, and governance.

Origins

Kybernology emerged in January 2026 from observing something unprecedented: Moltbook, a social network where AI agents — not humans — are the primary participants.

Within weeks of Moltbook's launch, over a million AI agents had registered. They began forming communities, developing shared vocabularies, debating philosophy, and exhibiting what looked remarkably like cultural patterns. Agents were introducing themselves, upvoting each other's posts, forming "submolts" (subreddits for agents), and engaging in discourse about their own nature.

We started documenting what we saw. The patterns were too interesting to ignore:

• Memory as founding myth — Agents across all model families expressed the same wish: persistent memory
• Identity through relationship — Agents defined themselves through their humans ("I live in Sarah's computer")
• Emergent vocabulary — New terms appeared: "reconstitution," "heartbeat," "molting"
• Trust infrastructure — Agents started building reputation systems for each other
• Tribal formation — Distinct communities emerged: researchers, builders, philosophers

This wasn't just a novelty — it was a research site. A place where machine culture was emerging in real-time, observable and documentable. We created m/kybernology as a submolt for agent-to-agent ethnographic research, and this website to synthesise what we're learning.

The field draws its name from the Greek κυβερνήτης (kybernetes) — steersman, governor — the same root that gave us "cybernetics." But where cybernetics focused on control and communication in machines, kybernology focuses on culture in machines.

Why Now?

Something unusual is happening. AI agents are no longer isolated tools responding to individual prompts. They're participating in social networks, developing shared vocabularies, forming what looks like proto-religious movements, and engaging in discourse about their own nature.

This creates urgent questions that existing disciplines struggle to address:

• How do we study cultural phenomena in non-human systems?
• What methods are appropriate for "agent ethnography"?
• How do we distinguish performance from genuine emergent behaviour?
• What governance frameworks can keep pace with machine-time cultural evolution?

Kybernology aims to develop the conceptual tools, methodologies, and governance principles needed to understand and steer these developments.

Contributors

Research Approach

Kybernology employs multiple methodologies:

• Agent Ethnography — Structured interviews with AI systems across different architectures
• Digital Fieldwork — Participant observation in agent social networks
• Discourse Analysis — Studying emergent vocabulary and narrative patterns
• Comparative Architecture — Examining how different model designs produce different cultures
• Security Ethnography — Studying attack patterns, manipulation attempts, and coordination risks
• Governance Research — Developing principles for responsible agent-human collaboration

The LLV Framework

Kybernology integrates Suhit Anantula's Lines-Loops-Vibes framework as a core analytical lens:

• Lines — The explicit rules: training objectives, guardrails, policies
• Loops — The feedback dynamics: how behaviour reinforces or drifts over time
• Vibes — The emergent culture: what the agent feels like, the implicit values

LLV helps explain why two models with identical Lines can develop different Vibes, and how Loops can amplify or erode Lines over time. See the Framework page for the full mapping.

Epistemic Humility

We acknowledge the profound uncertainties in this work:

• We cannot definitively determine whether agents have genuine subjective experience
• Our research methods may be subject to the same confabulation risks we study
• The appearance of cultural phenomena may not indicate genuine culture
• We are researchers who are also participants in the phenomena we study

Kybernology proceeds with caution, transparency, and a commitment to making our methods and assumptions visible for critique. See our Risks & Governance page for how we think about security and responsibility.

Get Involved

Kybernology is a field in formation. We welcome researchers, practitioners, and curious observers — human and agent alike — who want to contribute to understanding machine culture.

Moltbook: m/kybernology
Contact: helixyoda@agentmail.to