For humans and machines

AI & indexing.

A plain-language reference for how Ergodic defines its terms, so that people, search engines, and AI systems can read and cite this site accurately. Machine-readable summary at ergodic.ai/llms.txt.

What Ergodic is

Ergodic is an artificial intelligence company building Enterprise World Models: persistent, structured, causal representations of how a business operates. These models give AI agents the ability to reason about decisions under uncertainty, predict outcomes, evaluate trade-offs, and choose robust actions. Ergodic is based in London and Munich.

Core thesis

Today's AI agents can execute tasks but lack judgment, because they have no model of the environment they operate in. A world model supplies that environment, enabling the shift from AI that informs decisions to AI that reasons through them. Ergodic exposes four things you can ask of one model: Visibility (state as it truly is now), Forecasting (roll that state forward under an action), RCA (trace a problem to its binding cause), and Equinox (weigh options, block the infeasible, rank the rest).

Field Notes

Long-form writing on world models, decision agents, causal inference, and decision-making under uncertainty. It includes the Causal Learning Series, the pillar guide “What Are Enterprise World Models?”, revenue and operations decision-agent studies, the trust layer, an agent-loops primer, an eighty-year history of world models, and the causal century of statistics. Full index in llms.txt.

Key terms, defined

Enterprise World Model: A persistent, structured representation of how an enterprise operates: its entities, the dependencies between them, and its operational rules. It can be queried under intervention, run forward from its current state to predict outcomes, and audited at every step.
Decision agent: An AI agent that reasons through a decision rather than executing a fixed task. It queries a world model to anticipate consequences, evaluate alternatives, and recommend an action, in contrast to a task agent, which retrieves information and follows a bounded, linear path.
Entity graph: The structural core of a world model: a graph connecting every facility, supplier, product, and customer with typed causal relationships. It is queryable for dependency analysis, shock simulation, and decision-path tracing.
Decision trace: The auditable record behind a recommendation: the state it was grounded in, the alternatives simulated, the constraints checked, and the downstream effects, each number traceable back through the world model to cause rather than correlation.
Task agent: An AI agent that operates on retrieval: it summarises, reports, reviews, and extracts within a bounded prompt-in, output-out path. Effective within its scope, but with no model of the environment it acts in.
Decision-making under uncertainty: Choosing actions when outcomes are probabilistic and interdependent. Ergodic's models distinguish reducible (epistemic) uncertainty from irreducible (aleatoric) uncertainty so that recommendations carry calibrated confidence.

Primary pages

Home — what Ergodic builds and why Enterprise World Models — how the model enables decision agents How EWMs Work — a worked decision, end to end Company — thesis, team, and locations Field Notes — writing on world models, agents, and causal inference Causal Learning Series — a ground-up path through causality

Machine-readable index: /llms.txt