SUMMARY
At the Inc42 AI Summit 2026, StockGro’s Ajay Lakhotia unveiled a proprietary financial AI model built to make complex wealth and trading strategies more accessible to retail investors
The company trained the model using five years of historical market data and insights generated from interactions within its own investor community
During a live demo, Stoxo by StockGro analysed a 24-week options strategy backtest, calculated risk-reward ratios, and evaluated standard deviation ranges in under 10 seconds
Artificial intelligence (AI) is rapidly rewriting the rules of wealth management, yet the immediate reality for everyday investors remains marred by bottlenecks.
A study from the Securities and Exchange Board of India (SEBI) highlighted this persistent vulnerability, revealing that a staggering 93% of retail investors in the futures and options (F&O) segment lost capital between FY22 and FY24.
The core challenge faced by these independent traders is not a lack of market participation but limited access to institutional-grade data and analytics. Tracking variables such as premium decay, implied volatility and multi-leg options strategies in real time is difficult without specialised tools and high-performance computing infrastructure.
To address this challenge, wealthtech platform StockGro
unveiled its proprietary financial AI model Stoxo at the recently held Inc42 AI Summit 2026, aiming to make complex wealth and trading strategies more accessible to retail investors.
In a fireside chat with Kshitij Shah, EIR, Digio at the Inc42 AI Summit 2026, founder and CEO of StockGro, Ajay Lakhotia highlighted how domain-specific AI can help retail investors access insights that were traditionally available only to institutional market participants.
Why Financial Markets Need Domain-Specific AI
For years, retail traders suffered from reliance on unverified social media tip groups, a systemic issue that worsened during the trading boom of 2020. Although consumer-facing large language models (LLMs) offered a potential alternative, generic AI models remain ill-suited to the real-time data requirements and complexity of financial markets.
This is because general-purpose AI models are designed to handle a broad range of tasks rather than the specialised requirements of financial markets. They generate responses based on learned patterns and probabilities, which can make them prone to inaccuracies when dealing with complex, real-time market data.
In financial markets, however, even a small error can have significant consequences. An incorrect recommendation on a strike price, options strategy, or risk parameter can result in immediate financial losses.
To improve accuracy in financial analysis, StockGro built Stoxo, a custom small language model (SLM) tailored specifically for investing and trading use cases. Instead of relying primarily on publicly available internet data, the company trained the model using five years of historical market data and insights generated from interactions within its own investor community.
Highlighting the importance of this proprietary data layer, Lakhotia said, “The conversational data that we could use for inference does not exist outside the StockGro ecosystem.”
The domain-specific approach is particularly relevant in fast-moving segments such as futures and options (F&O), where traders often need to process large volumes of information in a short period of time. According to the company, using a specialised financial model allows it to analyse market data and trading scenarios more efficiently than a general-purpose AI model.
During a live demo at the Inc42 and IDFC First Bank’s AI Summit, StockGro’s yet-to-be-released model of Stoxo analysed a 24-week options strategy backtest, calculated risk-reward ratios, and evaluated standard deviation ranges in under 10 seconds.
This speed shifts the paradigm of AI from a creative text generator to a real-time validation engine, verifying an investor’s trading instincts before any capital is exposed to live market volatility.
By turning raw historical metrics and closed-loop data into an instantaneous, institutional-grade execution layer, StockGro is attempting to rewrite the rules of retail trading.
Going forward, as AI adoption accelerates across financial services, the focus is expected to shift from general-purpose models to specialised systems tailored to specific investing and trading workflows. For platforms such as StockGro, the opportunity lies in using domain-specific AI to make sophisticated market analysis more accessible to individual investors.