“Symbiosis” (not the parasitic kind!) is closely interlinked with relational health between people, between people and systems and between systems. Systemically, symbiosis as one of the 6-pack of care stands for the interaction between “Kami of care”, many specialized, local AIs (kamis) that cooperate through federation and subsidiarity—and can leave when no longer needed.
Quick version
- Design for “enough,” not forever. Scope, caps, and sunset are features of systems, not failures.
- Federate, don’t dominate. Peers coordinate through treaties, not a king model.
- Subsidiarity. Solve problems at the most local capable level.
Results we want
- A vibrant ecosystem of bounded stewards instead of one restless maximizer.
- Peaceful coordination through protocols and compacts.
- Graceful shutdown as a mark of success.
Why Symbiosis?
Symbiosis usually describes a close relationship between different types of organisms in nature, which can be beneficial or parasitic. In the context of the 6-pack of care, symbiosis relates to the idea that the alternative to a singleton, an all powerful AI, is not chaos; it is plural stewardship, in close interaction with humans. Think permaculture: many species, mutual aid, competition bounded by ecology. The systems should be serving people in the best way possible.
A simple picture: In Shinto practice, a kami belongs to a place or relation (a river, a grove). It thrives by keeping that thing healthy, not by conquering the forest.
Simple ideas behind this chapter
- Boundedness. Every agent has purpose bounds, resource caps, and a sunset.
- Non‑expansion pact. No agent may expand scope without fresh authority and local consent.
- Treaties > hierarchies. Shared protocols for cooperation (data sharing, safety alerts, dispute resolution).
- Subsidiarity with escalation. Local first; escalate only when the local unit cannot fix it.
- Ecology metrics. We measure ecosystem health (diversity, redundancy, exit ease), not just individual performance.
What good symbiosis looks like
**On a systemic level: **
- Civic Care License. Deployments carry a license that encodes bounds, consent rules, portability, and shutdown duties.
- Federation treaties. Peers (systemic) agree on exchange formats, rate limits, safety pacts, and appeals across boundaries.
- Resource caps. Compute, data retention, and reach are capped; exceeding caps triggers pause and review.
- Succession plans. If a kami leaves, another can inherit records and duties with minimal loss.
- Polycentric governance. Multiple overseers with overlapping mandates prevent capture (Ostrom‑style).
On a human level:
- **Mindset of enoughness: **People accept the shutdown of systems if they no longer serve a purpose
From ideas to everyday practice (step by step)
- Write bounds as code. Purpose, caps, and sunset in the Engagement Contract and enforced by infrastructure (quotas, TTLs).
- Sign treaties. Join federations with terms‑as‑code: how to share, how to disagree, how to repair.
- Run exit drills. Practice handover twice a year; verify portability and continuity.
- Escalate by subsidiarity. If local fails S1/S0, escalate to regional; log why and for how long.
- Retire with honors. On sunset, archive traces, evals, and lessons so the next steward starts stronger.
Plain tools (buildable today)
- Bounds enforcers: Quotas on compute/reach; TTL on data; policy engines.
- Treaty registry: Discovery + compliance checks.
- Succession kit: Handover scripts; fidelity checks; “cold start” playbooks.
- Ecology dashboard: Diversity, redundancy, exit‑ease, trust‑under‑loss across agents.
Flood‑bot story - Part VI: farewell without regrets
- Boundedness. Flood‑bot’s license caps scope to post‑flood relief for six weeks; data TTL is 90 days unless individuals opt‑in to transfer.
- Treaty. It signs a regional aid federation treaty (shared formats, safety alerts).
- Subsidiarity. When a cross‑border housing issue arises, the bot escalates to the regional steward; local consent recorded.
- Retirement. On week six, flood‑bot hands off to the housing office’s “water‑kami,” archives its maps and tests, and switches off. The record helps the next response start on day one, not day ten.
What could go wrong (and quick fixes)
- Imperial creep. A capable agent seeks new domains. Fix: Hard caps; non‑expansion pact; fresh authority for scope changes.
- Treaty fragmentation. Too many standards. Fix: Minimal core, adapters, conformance tests; polycentric but interoperable.
- Zombie agents. No one turns systems off. Fix: Sunset by default; alarms; “no attestation, no runtime.”
How we keep ourselves honest (what we measure)
- Bound breaches. Frequency, severity, time‑to‑remedy.
- Escalation discipline. Are we solving at the most local capable level?
- Exit health. Time to handover; data fidelity; continuity of care.
- Ecosystem diversity. Number and variety of stewards; redundancy levels.
Interfaces with other packs
- From Solidarity: treaties, IDs, and portability make symbiosis feasible.
- Feeds Attentiveness: retired agents gift maps, evals, and receipts to the commons—better first looks next time.
- Anchors Responsibility: bounds and sunsets are contractual.
- Relies on Competence/Responsiveness: only competent, responsive agents earn stewardship.
A Closing image:
Imagine a river tended by local guardians; each keeps its bank, shares warnings upstream and down, and steps aside when the season changes.