AI ENGINE
/ 07

Intelligence that can analyze, act and protect.

The EX AI Engine is a shared capability layer for analysis, market interaction and risk-aware decision support.

EX AI
AI ANALYSTSynthesizes signalCONCEPT
AI TRADERModels scenariosCONCEPT
AI RISKMonitors exposureCONCEPT
3Planned capabilities
VersionedModel lifecycle
HumanDecision control
HOW INTELLIGENCE IS DEVELOPED AND CONTROLLED

From research to monitored production—with humans in control.

The AI Engine is intended to support the full model lifecycle. Research, backtesting, deployment, explanation and drift monitoring sit inside one capability layer shared across EX products.

01

Model lifecycle

Research and feature development progress through historical evaluation, calibration, controlled deployment and ongoing performance monitoring.

  • Research and backtesting
  • Versioned deployment
  • Drift monitoring
02

Explainability at use

Interfaces should distinguish sourced facts from model interpretation and show the signals, update history and uncertainty behind a view.

  • Contributing signals
  • Update history
  • Source traceability
03

Clear agent boundaries

AI Analyst, AI Trader and AI Risk Engine describe decision-support capabilities—not autonomous financial actors.

  • Analysis and synthesis
  • Scenario support
  • Exposure monitoring
HOW IT WORKS

The governed model lifecycle

Consequential outputs move through evaluation and permission boundaries before they reach a product surface.

01
RESEARCH

Develop the model

Form hypotheses, features and evaluation criteria against historical data.

02
VALIDATE

Test calibration

Measure error, robustness and uncertainty across relevant conditions.

03
DEPLOY

Release with control

Version the model, define permissions and monitor production behavior.

04
REVIEW

Keep humans accountable

Require human approval for consequential actions and material changes.

EX PRINCIPLE / HOW INTELLIGENCE IS DEVELOPED AND CONTROLLED

AI supports the decision; it does not hide the decision.

EX product direction keeps sources, assumptions, uncertainty and approval boundaries visible.

WHY IT MATTERS
AI should make complex markets more legible while keeping humans in control of consequential decisions.
CORE CAPABILITIES

Three agents. Clear human boundaries.

01

AI Analyst

Synthesizes statistics, market movement and live context into concise, explainable intelligence.

An analyst should explain what changed, why the source is relevant and which conclusions remain tentative. A concise summary is useful when readers can inspect the evidence behind it and correct an unsupported interpretation.

  • Match briefings
  • Signal synthesis
  • Probability explanations
02

AI Trader

A future decision-support layer for strategy simulation, execution planning and market monitoring.

Simulation and execution are separate stages. A proposed strategy needs explicit assumptions, costs and constraints; any consequential action would require the appropriate user authorization and product controls. This website does not offer autonomous trading.

  • Scenario modeling
  • Execution support
  • Portfolio context
03

AI Risk Engine

Continuously evaluates exposure, concentration and changing conditions before they become hidden risk.

A warning should identify the affected exposure, the triggering condition and a review path. Human operators need to understand when a control is informative, when it restricts activity and how an exceptional case is escalated.

  • Exposure analysis
  • Anomaly detection
  • Adaptive controls
CONTINUE EXPLORINGEXT Ecosystem