SEBI Plans AI Guardrails for Capital Markets to Protect Investors

SEBI’s AI Rules Put Accountability Alongside Automation

The proposed framework aims to let capital-market participants use AI while keeping investor protection, human oversight and responsibility in view.

From broad principles to operational guardrails

The forthcoming guidelines will move SEBI’s AI and machine-learning agenda from proposed principles towards more detailed expectations for market participants. The regulator has said the aim is to capture the benefits of these technologies while limiting risks to investors, market integrity and financial stability. The precise requirements have not yet been set out.

That distinction matters. AI tools can inform or support decisions, but their use may also make it harder for investors to understand how an outcome was reached or who is responsible when something goes wrong. Disclosure requirements could help make the use of such systems more visible, while accountability provisions would keep responsibility with the regulated entity using them. These are potential effects of the proposed approach, not a description of final rules.

Human oversight is a signal about responsibility

SEBI chairman Tuhin Kanta Pandey has highlighted a “human-in-the-loop” approach, under which critical decisions remain subject to human oversight. This signals that the regulator does not view automation alone as a substitute for judgment and accountability in sensitive market functions.

For firms, the eventual challenge may be to show where AI is used, how decisions are reviewed and who can intervene when a system produces an outcome requiring scrutiny. For investors, clearer accountability could make it easier to identify the responsible institution. The guidelines’ practical impact will depend on how SEBI defines critical decisions and the level of oversight it expects.

AI supervision sits within a wider technology-risk agenda

The regulator’s focus extends beyond rules for market participants. SEBI has developed supervisory tools, including its “Sudarshan” tool, and is working to strengthen its AI capabilities through international initiatives and supervisory toolkits. These efforts point to a two-sided use of AI: firms may adopt it in market activities, while the regulator also develops tools to support supervision.

Cybersecurity is another part of this agenda. Pandey cited threats to financial institutions and cryptocurrency platforms, alongside digital fraud involving social media influencers. SEBI has introduced a structured incident-response mechanism aligned with the Financial Stability Board’s FIRE reporting template. The regulator is also inventorying cryptographic systems to assess vulnerabilities linked to quantum computing, with concerns expected to become more pronounced around 2032–35.

What investors and firms should watch next

The key test will be whether the final guidelines provide usable standards without making responsible innovation impractical. Broad expectations may leave firms uncertain about compliance; overly prescriptive rules could constrain the development of useful applications. SEBI’s emphasis on balancing innovation with safeguards suggests that implementation details will matter as much as the stated principles.

Investors should distinguish the regulator’s stated direction from obligations that are yet to be detailed. The June 2025 consultation paper provides the policy background, but the upcoming guidelines will determine how disclosures, accountability and human review are applied. Until then, the announcement is best understood as a signal of closer regulatory attention to both AI adoption and technology-related risks in capital markets.

Key Highlights

• SEBI plans detailed guidelines for responsible AI use in capital markets.
• The proposed approach includes disclosures, regulated-entity accountability and human oversight of critical decisions.
• A June 2025 consultation paper set out guiding principles for AI and machine-learning applications in securities markets.
• SEBI is building supervisory capabilities, including through its “Sudarshan” tool.
• Cyber incident response and potential quantum-related cryptographic vulnerabilities are also on the regulator’s agenda.

Investor Note: The proposed framework is not yet a final set of compliance requirements. Investors and market participants should watch for the detailed guidelines, which will clarify how SEBI intends to balance AI-led innovation with transparency, human judgment and accountability.