Last updated: March 22, 2026
1. Overview
Fortunato AI, Inc. is committed to transparency about how we use artificial intelligence across our products. This document explains the AI systems we employ, their capabilities, their limitations, and how we work to ensure responsible use.
2. AI Systems in Use
QuorumTrade — Trading Intelligence
QuorumTrade employs a multi-agent AI system consisting of 18 specialized agents organized in a 9-layer pipeline:
- Regime Detection Agents (L3): Classify market conditions (trend, range, volatile, crash) using technical indicators and statistical models
- Strategy Agents (L4): Generate directional signals using trend-following, mean-reversion, momentum, breakout, and ML-based strategies
- Deliberation Agents (L5): Aggregate signals, compute Trade Quality Scores, and apply regime-adaptive thresholds
- Risk Agents (L6): Size positions, enforce risk limits, and apply governor constraints
- Execution Agents (L8): Route orders, manage stop-losses, and track positions
- Learning Agent (L9): Adjusts agent weights based on historical trade outcomes
OpenClaw — AI Enrichment
15 specialized AI agents providing research, sentiment analysis, fundamental analysis, compliance checks, and macro context. These agents use large language models (LLMs) including GPT-4 and Claude for natural language analysis and reasoning.
Growth OS — Growth Automation
AI agents for lead generation, outreach optimization, and pipeline management. Uses LLMs for content generation and prospect research.
3. Model Transparency
Foundation Models Used
- OpenAI GPT-4: Used for research, analysis, and natural language tasks in OpenClaw agents
- Anthropic Claude: Used for complex reasoning, compliance analysis, and content generation
- Custom Statistical Models: Proprietary models for regime detection, signal generation, and risk calculation (not LLM-based)
Data Sources
- Real-time price feeds from cryptocurrency and equity exchanges
- Orderbook depth data
- Technical indicators computed from OHLCV data
- Sentiment data from public sources
- On-chain metrics for cryptocurrency assets
- Public financial data and market news
4. Limitations
Our AI systems have significant limitations that users should understand:
- No Prediction Guarantee: AI cannot predict future market movements with certainty
- Historical Bias: Models are trained on historical data that may not reflect future conditions
- Hallucination Risk: LLM-based agents may generate plausible but incorrect analysis
- Regime Lag: Market regime detection may lag actual regime changes
- Data Dependency: System quality is limited by input data quality
- Black Swan Events: AI cannot anticipate unprecedented market events
- Correlation Assumptions: Correlation models may break down during market stress
5. Human Oversight
While our trading agents operate autonomously within the paper trading simulation, we maintain human oversight through:
- AUTO_TRADE flag enforced as False — no real trading can occur
- Risk governor with hard limits on exposure, position sizing, and drawdown
- Telegram alerts for all significant events
- Comprehensive event logging for audit and review
- Manual override capabilities for all system parameters
6. AI Ethics Principles
We follow these principles in our AI development:
- Transparency: We disclose how AI is used and what its limitations are
- No Deception: We never misrepresent AI capabilities or simulated results
- User Safety: Paper trading only — we do not put user capital at risk
- Data Privacy: AI systems do not use personal user data for model training
- Accountability: All AI decisions are logged and auditable
7. Updates
We will update this disclosure as our AI systems evolve. Material changes will be communicated to users.
8. Governing Law
This AI Disclosure shall be governed by the laws of the State of Delaware, United States.
9. Contact
For questions about our AI systems:
Fortunato AI, Inc.
Email: [email protected]