Six Autonomy Framework

The Six Autonomy Framework defines autonomy as freedom from physical, supply chain, and operational constraints — not as a software feature. Developed by ElectronsX, it maps the six dependency layers that determine whether an industrial system, fleet, or facility can sustain operations when upstream constraints fail, tighten, or concentrate.

The layers are sequential because each one constrains the next. Materials affect what silicon can be built. Silicon affects what compute and control systems exist. Energy affects operational uptime. Heat affects throughput density. Data affects sovereignty of decision-making. Operational autonomy — the ability to run continuously without routine human presence — is the end state produced when all five preceding layers are resolved.

Six Autonomy Layers

Layer	Primary Constraint Removed	What Breaks Without It	Scope
1. Materials Autonomy	Critical materials concentration — battery metals, REEs, copper, graphite, GOES steel.	Production slows, substitution costs spike, deployment cadence collapses.	Mining, refining, recycling, diversified sourcing, inventory buffering.
2. Silicon Autonomy	Semiconductor fabrication concentration — logic, memory, power, automotive MCUs.	Controllers, power electronics, vehicles, robots, AI systems become supply-constrained.	SiC, GaN, HBM, logic fabs, substitute architectures, diversified sourcing.
3. Energy Autonomy	Grid dependency — capacity constraints, price volatility, outage exposure.	Uptime, expansion, throughput, and resilience capped by grid weakness.	Microgrids, BESS, on-site generation, islanding, demand shaping.
4. Thermal Autonomy	Heat removal limits — cooling bottlenecks, thermal density ceilings.	Throughput drops, throttling rises, rack density stalls, battery performance degrades.	Liquid cooling, immersion, heat reuse, facility thermal zoning.
5. Data Autonomy	Centralized AI inference dependency — proprietary models, opaque weights, third-party pipelines.	Critical decisions depend on external uptime, policies, and non-auditable training.	Local inference, auditable models, federated deployment, protected operational data.
6. Operational Autonomy	Human presence dependency — routine supervision, control, execution.	Facilities remain labor-bound, shift-bound, exception-heavy.	Lights-out facilities, autonomous fleets, remote supervision, self-optimizing operations.

Framing

Autonomy is often framed as a software feature, but in real systems it is a stack problem. A robot, fleet, gigafactory, mine, port, or AI data center cannot be autonomous if it is blocked by concentrated materials, delayed by chip shortages, power-limited by the grid, thermally throttled by heat rejection, dependent on cloud inference, or unable to function without humans on-site.

The practical meaning of autonomy is freedom from operational choke points. The six layers define those choke points in order of strategic importance.

The Three Walls

The Three Walls refers to the most visible hard constraints in the modern autonomy stack: energy, silicon, and heat. The Six Autonomy Framework expands this into a full dependency model by adding upstream materials risk, downstream data sovereignty, and end-state operational execution.

Three Walls Concept	Corresponding Autonomy Layer	Expanded Interpretation
Silicon Wall	Silicon Autonomy	Moves from chip scarcity framing to sovereign sourcing, substitution, buffering, and geographic fabrication resilience.
Energy Wall	Energy Autonomy	Moves from power shortfall framing to microgrid-native resilience, local generation, storage, and dispatch intelligence.
Thermal Wall	Thermal Autonomy	Moves from cooling bottlenecks to full-system freedom from heat rejection limits across compute, batteries, power electronics, and facilities.

Materials Autonomy sits upstream of the Three Walls. Data Autonomy and Operational Autonomy sit downstream. Together, the six-layer model explains not just where systems hit limits, but how to remove those limits in sequence.

Readiness Bands

Each autonomy layer has its own readiness band framework with layer-specific criteria. See individual overview pages: Materials Autonomy, Silicon Autonomy, Energy Autonomy, Thermal Autonomy, Data Autonomy, Operational Autonomy.

Band	General Meaning	Typical Characteristics
A-0 Dependent	The system is structurally dependent on an external constraint.	Single-point exposure, opaque dependencies, weak buffering, minimal local control.
A-1 Aware	The dependency is known and partially mapped, but not materially reduced.	Risk identified, pilot mitigations in place, dependency remains primary.
A-2 Hybrid	The system can partially operate around the constraint using alternate pathways.	Diversified sourcing, partial local control, fallback modes, documented flows, selective resilience.
A-3 Autonomous	The primary operational dependency has been removed for critical functions.	Controlled core stack, resilient architecture, auditable operation, independent execution for mission-critical workloads.

Tesla Case Study

Tesla represents the most complete real-world implementation of the Six Autonomy Framework by a single organization. Materials Autonomy: lithium refinery in Corpus Christi, LFP cell partnership with LG Energy Solution in Michigan, MP Materials relationship for NdFeB magnets. Silicon Autonomy: Samsung Taylor sole-customer arrangement for AI6 chips, Terafab captive semiconductor facility breaking ground March 2026. Energy Autonomy: Megapack, Solar Roof, Autobidder, GigaTexas microgrid. Thermal Autonomy: proprietary battery and power electronics thermal management, Cortex 2.0 cooling at 500 MW scale. Data Autonomy: Dojo captive training cluster, closed fleet training data, on-vehicle FSD inference. Operational Autonomy: Optimus deploying in GigaTexas, Cybercab launching in Austin.

No other organization has simultaneously addressed all six layers. Tesla's stack demonstrates that the dependency sequence is real — each layer required investment before the next became tractable.

Closing Perspective

The sequence is not arbitrary. Materials precede silicon because chips require refined inputs. Silicon precedes energy because control systems require chips. Energy precedes thermal because heat is generated by energy consumption. Thermal precedes data because compute requires cooling. Data precedes operational because autonomous decisions require local inference. Remove any layer and the layers above it become fragile."

The next phase of industrial and AI deployment will not be defined by software alone. It will be defined by who can remove physical, thermal, electrical, informational, and operational dependencies in the correct order.

---