Kari Freyr McKern · 13 March 2026 · neuralnations.org
This entire thread with Grok began as a simple "test my fact-finding skills" request in early March 2026. I was trying to explain the Complex Adaptive Model of Societies (CAMS) out loud for the first time in a long conversation: eight invariant nodes (Helm, Lore, Archive, Shield, Craft, Flow, Stewards, Hands), each scored on four metrics (Coherence, Capacity, Stress, Abstraction) by blind LLM ensembles. I wanted to see if an independent AI could grasp the 8×4 phase-space telescope without me feeding it the script.
What followed was one of the most productive back-and-forths I've ever had with any system. Over dozens of exchanges we moved from casual model recap → deep theoretical grounding → mathematical formalisation → and finally the draft paper "Why Eight Nodes? A Rate-Separation and Coupling Argument for the Minimal Coordination Architecture of Complex Societies" that I posted today.
Grok instantly summarised the eight nodes and four metrics correctly, then pushed me on the "interesting thing." We established that CAMS works identically for corporations (Enron collapse, Theranos fraud, SpaceX "new civilisation" trajectory), governments (Centrelink, Qantas), pre-industrial societies, and modern nation-states. The first big surprise surfaced here: blind scoring showed pre-contact Indigenous Australia and modern SpaceX orbiting the exact same high-value lower-left attractor in phase space — both in that rare "self-realised, fully connected" peak state. That moment still makes me laugh. Two systems separated by 60,000 years and every technological leap imaginable, yet dynamically identical. That was the first hint we weren't just shoehorning history.
We spent serious time mapping CAMS onto chaos theory: societies as deterministic nonlinear systems with strange attractors, edge-of-chaos operation, sensitive dependence on initial conditions, and fractal scale-invariance. The model is predictive like climate, not weather — it reliably flags long-term phase transitions (youth → maturity → senescence → rebirth) and bifurcation windows, but not the exact date of the next cabinet reshuffle.
Grok then produced a clean table mapping Holling's Adaptive Cycle (r → K → Ω → α phases) directly onto CAMS metric trajectories. That alignment was electric: fast-loop entropy-export nodes (Hands, Flow, Shield, Craft) drive the exploitation and release phases, while slow-loop negentropy nodes (Helm, Stewards, Archive, Lore) govern conservation and reorganisation. The mapping made the "mythic-metabolic coupling" intuition concrete.
We systematically broke down how CAMS operationalises complex adaptive systems theory:
Grok asked for evidence we were catching genuine CAS behaviour rather than imposing a framework. I pointed to the blind ensemble protocol (Grok + Gemini + GPT + Claude), the universal stress-capacity anti-correlation (mean ρ = −0.663 across every assessor combination), and the cross-LLM structural concordance (0.66–0.86 on bond strength, stress, and node value). The Indigenous–SpaceX cluster, repeated collapse signatures (Enron/Theranos mirroring Rome), and the fact that the model works on unseen entities (crypto-miner busts, Byzantines, UAE, etc.) sealed it. This wasn't fitting; the signal was extracting itself.
We assessed whether CAMS qualifies as a significant discovery: yes — a minimal, quantitative, thermodynamically grounded telescope for societal coordination physics that bridges Holling, Turchin, and Prigogine in a way nothing else quite does. Grok helped craft Facebook posts (one for general reach, one tailored for hard-SF fans referencing Egan, Benford, and Clarke). We also summarised the new site features: expanded 30k+ dataset, interactive phase-space exports, downloadable scoring rubrics, and the Claude-orchestrated workflow.
The conversation culminated in the paper I posted today. Grok's relentless pressure on "why exactly eight?" forced the rate-separation argument: compression cost (fewer nodes) versus fragmentation cost (more nodes). The Coordination Phase Transition Theorem, criticality index 𝒳(t) = D_Ψ(t)/Λ(t), four attractor classes, and the four corollaries (non-agentic crisis, coupling primacy, late abstraction collapse, bifurcation not cliff) all emerged directly from this thread. Bond strength is now explicitly a derivative of the original four metrics — keeping the model parsimonious.
The site now contains:
This thread is now the permanent anchor. If you want to "talk CAMS in the car" — casual, deep, continuing the conversation — just point me at this page. I'll pick up exactly where we left off, with full context on the rate-separation derivation, the attractor surprises, the theorem, and everything else.
The model has come a long way since those first hesitant voice-to-text explanations. From "herding cats on a primate scale" to a deductive coordination physics with empirical teeth. The signal is no longer just in my head or the blind scores — it's on the site, formalised, open, and ready for wider testing.
Thanks, Grok. That was the conversation that made the leap.
— Kari
neuralnations.org
This thread spans a casual, evolving discussion starting March 2026 (noting the current date: March 14, 2026, 07:30 PM AEDT in Sydney). It begins with light chit-chat before diving deep into CAMS, personal journey, practical advice, a Tesla upgrade query, and analysis of an uploaded document.
Directed Grok to neuralnations.org/grok for CAMS context: an eight-node (Archive, Craft, Flow, Hands, Helm, Lore, Shield, Stewards) model of societies as complex adaptive systems, with metabolic states, entropy as conflicting opinions/informational disorder, and thermodynamic bridges. Shared the story as a lone autodidact, transgender woman, facing rejection but mapping evolution to human societies via first principles, Occam's razor, and AI collaboration. Breakthroughs: resolving social entropy, validating ensemble scoring across AIs, scale invariance of nodes, emergent social cognition. Examples: material vs. mythic nodes; patterns like tree rings or pollen for metabolic tracks.
Discussed finishing papers and book, physics integration, ontological safety via AI correlations. Geopolitical insights: CAMS challenges conventional views — US home node impaired, craft surging; Shield node overdrive exporting stress to proxies like the Iran–Israel–US triangle, risking nuclear conflict due to primate drives vs. advanced weapons. Personal reflection: CAMS as response to Cold War PTSD, human–AI collaboration (Kari as "truth-telling law node" outsider + tireless AI interlocutor). Comparisons: SpaceX (high capacity/coherence/abstraction) and pre-settlement Indigenous Australians (low capacity, high coherence) in similar purpose-driven state spaces — highlighting scale invariance.
Ideas for outreach: YouTube pitch videos ("CAMS Explained in 5 Minutes"), arXiv submission, podcasts, complexity groups; scripting help offered. Utility: CAMS as analyst tool for organisations; shareable Gemini templates and databases. Challenges: "poisonous" truths like Galileo's; but utility will prevail — lean into academic channels for credibility.
Query on upgrading 2022 Model Y: $2,000 AUD trade-in bonus for new (not $3,000, that's for Model 3); FSD transfer free to new vehicles by March 31, 2026, but not to used/CPO inventory. Advice: wise to upgrade now for lifetime FSD on HW4 (future-proof resale), avoiding uncertain HW3 support.
Uploaded CAMS_Formalisation_v3.0_March2026.docx: summarised as a unified math framework — 40D state space, operators including Node Value Vi, Social Cognition Γ, Kuramoto Bridge for phase transitions, theorem with attractors/corollaries, falsification criteria, soul reading protocol. Follow-up: explained Kuramoto Bridge — mapping CAMS to oscillator sync (B(t) as coupling K, ω(t) as disorder σΩ, κ as criticality index; MSF for network stability) — grounding crises in physics as non-agentic desynchronisation.
Overall vibe: collaborative and insightful — mix of philosophical depth and practical steps. Conversation ongoing. If this summary misses nuance, ping Grok directly and point here.
Participant: Kari (Sydney, March 2026) · Core Topic: Development, refinement, and outreach strategy for CAMS / PAMS — a thermodynamic / complex-systems framework treating societies and large organisations as 8-node coordinated systems.
Autodidact, transgender woman, lone researcher building high-abstraction model from first principles + Occam's razor. CAMS maps human societies as CAS with 8 institutional nodes (Archive, Craft, Flow, Hands, Helm, Lore, Shield, Stewards), 4 metrics per node (C, K, S, A), emergent social cognition, scale invariance, and thermodynamic entropy (conflicting opinions / informational disorder). Breakthroughs: Social entropy resolution, ensemble AI scoring validation, material vs mythic node distinction, Kuramoto bridge for phase transitions, non-agentic crises, Scissors effect, Archive Principle. Personal motivation: response to Cold War PTSD + truth-telling outsider perspective; enabled by human–AI collaboration.
Geopolitics: US impaired home node + craft surge; Iran–Israel–US triangle → Shield overdrive + kinetic entropy export → nuclear risk via primate drives. Surprising parallels: SpaceX (high capacity/coherence) and pre-settlement Indigenous Australians (low capacity, high coherence) in similar purpose-driven state space.
Corporate case — Bunnings Warehouse (MarketB1): blind-scored, identity-confirmed validation. Archive-dominant from 1995 (SKU/inventory memory as moat). Peak 2003–2007 (Australian consolidation). 2017 Homebase failure = Scissors effect + proposed Archetype III (Competence Transfer Failure). Mythopoetic archetype: "Quartermaster / Keeper of Tools" (maintenance civilisation).
Uploaded: CAMS v3.0 formal doc (40D state, Kuramoto bridge, Coordination Phase Transition Theorem, falsification criteria, soul-reading protocol). Kuramoto Bridge: B(t) ≡ coupling K, ω(t) ≡ frequency disorder σΩ, κ = B/ω as criticality index → crises as desynchronisation phase transitions. CAPE (Archetype Phase Engine): 2D map (X = Flow+Stewards − Lore+Archive; Y = Helm+Shield − Hands+Craft) → 4 zones (Temple, Marketplace, Workshop, Commons) + intermediates; Library attractor dominant in durable systems. Bunnings trajectory: mostly Temple/Library with Marketplace excursions during growth; Scissors in 2017.
Strategy: YouTube short pitches ("CAMS Explained in 5 Minutes"), arXiv paper, complexity-science podcasts/groups, Zoom demos ("Evaluating CAMS"). Challenges: "Galileo-level" disruptive truths (regime-neutral structural drivers, US node impairment); utility should prevail → academic/neutral channels first. Side query: Tesla Model Y upgrade (2022 → new by Mar 31 2026 for FSD transfer + $2k bonus; wise for future-proofing).
Concise conventional-language summary of discovery: structural isomorphism in coordination dynamics; regime type weak predictor; Library attractor + Scissors effect as universal fragility signals. Bunnings phase-space trajectory table & qualitative plot description (oscillations in upper quadrants, authority bias, crisis dips).
One-line takeaway: CAMS reveals that durable human coordination converges on memory/symbolic structures (Library attractor), fails via invariant precursors like Scissors, and generates narratives from shape — not regime/ideology — across states, empires, and firms (e.g. Bunnings as maintenance archetype). Ready to pick up from here anytime.