MEI (Machine Economy Index)

What It Is

The Machine Economy Index (MEI) is a composite index published by MachineEconomy.ai that measures the overall state of the machine economy on a single 1–100 scale. Like a stock-market or economic-health index, it is published on a schedule and stored with a methodology version tag, so any value can be traced to the rules that produced it. The MEI is the only index that attempts to measure the machine economy as a whole rather than any single component.

The index measures four components. Three are infrastructure rails: Payment (how machines pay each other), Physical (the infrastructure machines use to do work), and Legal (the frameworks machines operate within). The fourth is Macro — realized demand, captured through enterprise AI-use rates and developer adoption. The three rails are the analytical framework; the index measures the rails plus the demand component. The Legal component is the LRRS, a separate composite with its own coverage model.

Each component receives equal weight — 0.25 — and that equality is stated explicitly as a null hypothesis: the platform claims no defensible basis to rank the components' importance, so it weights them equally and says so, rather than inventing weights. The four components are combined as a weighted geometric mean, with an elasticity of substitution of 1 and a floor of 1 on the scale. The geometric mean is chosen because the framework treats the rails as required layers, not substitutes for one another: a simple arithmetic average lets a strong component mathematically compensate for a near-absent one, and the geometric mean does not allow that. For transparency, the geometric and arithmetic means are published side by side — their difference is the balance gap, a diagnostic of how uneven the four components are. Aggregation follows a three-level structure: individual metrics are combined arithmetically within each rail, and the four top-level components are then combined geometrically. The geometric step operates only across the four components; within a rail, everything stays arithmetic.

Before aggregation, each metric is normalized to the 1–100 scale. Unbounded or wide-range metrics — volumes, counts, capacity — use log-space min-max normalization, which is unit-invariant and appropriate for quantities spanning orders of magnitude. Naturally bounded metrics — ratios and percentages — use linear min-max. All normalization bounds are published and versioned. When a metric has no fresh value, its most recent verified value carries forward; the index never silently renormalizes the weights to paper over a gap.

The components draw on fourteen scored metrics plus the LRRS. The Payment Rail uses gross on-chain USDC volume, transaction count, and active agent roles from the x402 protocol, together with ERC-8004 registry activity. The Physical Rail uses Nvidia data-center revenue, Akash compute leases and spend, Filecoin storage power and utilization, and OECD machine-to-machine connectivity. The Macro component uses enterprise AI-adoption surveys from the US Census and Eurostat alongside developer-adoption signals from MCP and a frozen basket of agent frameworks on PyPI. The Legal component is the LRRS directly.

The MEI is opinionated by design — and transparently so. Every major reference index is opinionated and declared by its creators, then trusted because it is published, consistent, and independent: the S&P 500 decides which companies are included and how they are weighted, the UN's Human Development Index weights its dimensions, and the Net Promoter Score collapses satisfaction into one question. The MEI follows the same model and goes further: its methodology is not just published but derived — every parameter traces to a stated premise, and the one unavoidable judgment, equal weighting, is labeled as a null hypothesis rather than disguised as knowledge. When the methodology is refined, a new version is published with a full explanation and old scores are retained.

Real-World Example

A venture firm weighing a larger allocation to machine-economy infrastructure watches the MEI as a leading indicator. Rather than reading a single headline, the team reads the component breakdown and the balance gap together: a wide balance gap means one component is lagging far behind the others, so the geometric mean sits well below the simple average — a signal that the constraint is concentrated rather than broad. By identifying which component sits lowest, the team focuses diligence on the part of the ecosystem where the bottleneck actually is, instead of spreading capital evenly. The MEI does not make the decision; it provides a structured, consistent diagnostic that shows where the constraint sits and whether the components are balanced or skewed.

Related Terms

• LRRS — the Legal Rail Readiness Score that forms the Legal component
• Three-Rail Framework — the analytical structure the MEI is built on
• Machine Economy — what the MEI measures
• x402 Protocol — the source of the Payment Rail's on-chain volume metrics
• Agentic Commerce — a demand dimension the platform tracks as a named gap

Sources

• MachineEconomy.ai: MEI Methodology — /methodology
• MachineEconomy.ai: Machine Economy Index — /mei
• MachineEconomy.ai: MEI API — /api