Zoonova AI announced the launch of Alpha AI, a new investing platform designed to make advanced market intelligence more accessible through a natural-language AI Command Centre.
Alpha AI combines quantitative machine learning, technical pattern recognition, sentiment analysis, visual analytics, and guided research workflows in one interface. The platform helps users move from a simple stock question to a structured view of forecasts, risk, and market context.
At the core of Alpha AI is Zoonova AI’s Quad-Ensemble machine learning framework, which combines XGBoost, Random Forest, CatBoost, and Temporal Fusion Transformer (TFT) models. The framework uses hyperparameter tuning and RMSE-based optimisation to support predictive performance across multiple market conditions and time horizons. The system processes over 150 financial features using approximately 3 to 4 years of daily historical data per stock, retraining models weekly and updating core calculations twice daily.
Alpha AI also incorporates additional modelling layers. Zoonova AI uses a Birch model for pattern recognition across 200+ charts and technical signals and a VADER-based sentiment engine that processes approximately 3,000 live news feeds to generate stock-level sentiment. Gemini 3.1 Flash Lite helps translate quantitative outputs into clearer natural-language explanations.
Through the AI Command Centre, users can ask questions in plain English and generate multi-horizon alpha and price forecasts, tear sheets, Monte Carlo simulations, factor analysis, and stress tests. The platform presents complex analytics in a more approachable format for individual investors, traders, and other market participants.
Alpha AI also emphasises usability. The interface can generate up to 23 guided follow-up prompts spanning Deep Research reports, valuation and multiples analysis, comparative analysis, growth and moat analysis, risk registers, peer assessments, EPS forecasting, event calendars, and market-news summaries. Users can also use an “Explain this tear sheet” workflow to translate complex factor-analysis, Monte Carlo, and stress-test outputs into plain-English strategy, while in-line glossaries define technical terms directly inside the experience.