Economic security

A market can be technically resolvable and still be economically unsafe. The Economic Security Engine evaluates whether a proposed question should be opened and how much exposure it may accept.

Inputs

Risk evaluation can consider:

• asset liquidity and estimated manipulation cost;
• distance from a proposed threshold;
• observation-window length;
• source latency, reliability and finality;
• expected market concentration;
• single-market, single-asset and global protocol exposure;
• correlation with existing markets;
• duplicate or near-duplicate questions;
• evidence freshness and source diversity.

Decisions

The engine produces a structured decision such as:

• SAFE — eligible within an explicit exposure limit;
• UNSAFE — understood but outside the accepted risk boundary;
• UNSUPPORTED — missing a model, source or settlement capability.

Decisions include versioned reasons. A high popularity score or model confidence cannot override an unsafe result.

Fail-closed behavior

Missing required values do not default to permissive assumptions. A market is not created when the engine cannot establish the data, evidence path or economic conditions required by its template.

Runtime controls

Risk controls continue after market creation through:

• outcome and market exposure caps;
• idempotent position submission;
• entry-deadline enforcement;
• settlement quorum and evidence hashes;
• emergency controls with explicit scope and audit trails.

Emergency control must not become an undocumented ability to choose a winning outcome.