Context:
In a timely and forward-looking move, the Securities and Exchange Board of India (SEBI) has released a discussion paper titled “Guidelines for Responsible Usage of AI/ML in the Indian Securities Market.” This consultation initiative aims to regulate the rapid rise of artificial intelligence (AI) and machine learning (ML) applications—particularly algorithmic trading—on Dalal Street.
Why It Matters?
While AI doesn’t introduce fundamentally new risks to markets, it can amplify existing vulnerabilities such as market manipulation, flash crashes, systemic contagion, and biased decision-making. SEBI’s proactive stance is meant to ensure investor protection, financial stability, and ethical use of AI technologies in securities trading.
Key Elements of SEBI’s Proposed Framework
- Model Governance and Testing:
- Mandate for rigorous pre-deployment testing and periodic audits of AI/ML models.
- Explainability and traceability of model decisions to be ensured.
- Bias, Fairness, and Privacy:
- Guidelines on mitigating algorithmic bias, protecting investor data, and ensuring model integrity.
- Disclosure Requirements:
- Entities must disclose the nature of AI/ML systems used, data sources, model purpose, and decision-making logic to SEBI.
- Third-Party and Non-Traditional Players:
- Regulatory purview to expand to cover third-party algo vendors, fintech startups, and non-registered intermediaries using AI for financial services.
- Investor Protection and Oversight:
- Emphasis on human-in-the-loop supervision, especially as markets inch closer to agentic AI—AI systems that can act autonomously in investing.
Implications
- Market participants will need to revisit their AI governance frameworks, particularly around model validation and compliance reporting.
- Startups and fintechs must prepare for regulatory scrutiny if offering AI-based trading or advisory services.
- Investors may benefit from more transparent and explainable AI tools, subject to new disclosures and testing standards.
