SEBI deploys AI to monitor social media as 62% of investors trust finfluencers

Project Sudarsan: How Sebi is using AI to police finfluencers with 60% of investors trusting their advice

Sebi is stepping up its use of artificial intelligence to monitor misleading financial advice on social media. Its Project Sudarsan tracks unsolicited advice, while Sebi R(AI)DAR reviews advertisements. The regulator said 62% of investors are influenced by finfluencers, highlighting the need for stronger digital surveillance and investor protection. Markets regulator Sebi is using artificial intelligence and data analytics to track misleading financial advice on social media, as a new investor survey showed that 62% of investors are influenced by finfluencers. In its annual report, the market regulator said trust in the digital era can no longer be protected only through exchanges, clearing corporations and depositories. "Data has become a second layer of market infrastructure, making the quality of market data, the integrity of data systems and governance of data use central to investor protection," it said. The regulator said it has responded by investing in technology and data analytics as core supervisory tools, so that the investor protection framework scales along with the growth of the market. A key part of this digital push is aimed at unregistered financial influencers, many of whom operate on social media without accountability or verified performance records. Sebi said its latest investor survey showed that 62% of investors are influenced by finfluencers, creating the need for stronger digital vigilance. Project Sudarsan to track online advice Sebi said it has launched Project Sudarsan, a tool developed to monitor unsolicited financial advice on social media. It has also rolled out Sebi R(AI)DAR, an AI-enabled platform to review advertisements. Also Read: Info Edge Q1 Results: Standalone Profit falls 6% YoY to Rs 245 crore The regulator said these tools will help it identify unauthorised digital activity and finfluencers who may mislead investors through unverified claims. The action comes after a sharp rise in retail participation since the pandemic, especially in high-risk areas such as options trading. Sebi chairman Tuhin Kanta Pandey earlier said that several retail investors were being influenced by such online personalities to enter the risk-prone derivatives market, often through claims that large money can be made from trading. Sebi has already removed more than 1.2 lakh misleading social media posts by unregistered finfluencers and is using AI tools to track violations in the digital space. Fake apps also under watch Sebi's digital investor protection plan also covers payment verification and trading apps. The regulator has introduced Validated UPI handles and the Sebi Check facility, which allow investors to verify in real time whether a payment is going to a genuine Sebi-registered intermediary. It has also partnered with Google Play for a verified app label initiative. This will give investors a visible signal that a stock trading app belongs to a genuine Sebi-registered broker. The move is aimed at tackling fake trading apps, fraudulent payment requests and impersonation of registered intermediaries. Pandey had earlier said Sebi's action against finfluencers is not a heavy-handed crackdown. He described it as a calibrated exercise aimed at identifying problem areas and dealing with them. "Market development is not about a sledgehammer approach but more like a surgeon's knife -- identifying problem areas and dealing with them," he had said. (Disclaimer: Recommendations, suggestions, views and opinions given by the experts are their own. These do not represent the views of Economic Times)

SEBI is using AI to monitor social media for fraud

* Access the chapter nine of the SEBI Annual Report 2025-26 from here. The Securities and Exchange Board of India (SEBI) is using an AI system which scans public social media content for fraudulent investment claims. The system also detects unregistered advice and impersonation of regulated entities. SEBI disclosed this in its Annual Report 2025-26. The regulator said Project SUDARSAN became operational in November 2025. It has identified more than 20,000 instances of fraudulent content and posts on social media. How SUDARSAN Works: SUDARSAN stands for Surveillance of Unauthorized Digital Activity via Real-time Scanner for Anti-fraud. It is an AI powered intelligence system that monitors publicly available videos, messages, images and advertisements. The system scans content across major social media platforms. SEBI says the system uses multimodal AI to analyse speech, visuals, regional-language content and the context in which claims are made. It combines this information with behavioural and regulatory parameters, assigns risk scores and produces alerts for further scrutiny. SUDARSAN is designed to detect promises of guaranteed investment returns and fake certifications. It also detects impersonation of SEBI-regulated entities. The system flags investment advice offered without the required registration. The concern comes against the backdrop of SEBI's Investor Survey 2025. According to reports citing the survey, 62% of investors said finfluencers affect at least some of their investment decisions. Meanwhile, 93% considered financial influencers moderately to highly credible. Among those surveyed, people used YouTube most often for securities-market information, followed by Instagram and Facebook. SEBI Also Deploys R(AI)DAR: SEBI has separately deployed R(AI)DAR, an AI-based tool that reviews advertisements and investor education material issued by asset management companies. The system flags possible violations, including missing mandatory disclosures and the promotion of investment products as educational content. The Association of Mutual Funds in India has deployed the system. SEBI describes these systems as part of a shift from manual and complaint-led supervision. The regulator is moving towards technology-based monitoring. It aims to detect potentially harmful content earlier. The annual report does not explain how SUDARSAN calculates its risk scores, what thresholds trigger regulatory scrutiny, how it handles false positives, or what safeguards it uses to govern the collection and retention of social media data. Earlier Social Media Monitoring Plans Drew Scrutiny: The use of automated tools to monitor online activity, however, has a longer and contested history in India. In 2018, the Union government proposed a Social Media Communication Hub to collect and analyse online conversations and track trends, influencers and individual accounts. The tender also referred to monitoring platforms including WhatsApp, Twitter and Instagram, as well as emails. During a challenge to the project, the Supreme Court observed: "The government wants to tap citizens' WhatsApp messages. It will be like creating a surveillance state." The Centre subsequently withdrew the proposal and told the court that it would review its social media policy. A similar controversy emerged in 2020. Government-owned Broadcast Engineering Consultants India Limited sought tools for fact verification and disinformation detection. The tender included monitoring people who uploaded disinformation. It also sought to identify influencers and their locations. The tools would map links between entities. They would also use AI and machine learning to verify claims. The Internet Freedom Foundation challenged that proposal, arguing that such automated monitoring could enable mass surveillance and affect privacy and free speech. Why it matters: SEBI says SUDARSAN only uses publicly available content linked to securities-market fraud and regulatory violations. However, its annual report does not explain which social media platforms it covers. It also does not state how long it retains the collected data. The report does not explain the safeguards for automated risk scoring. It also does not specify what recourse users have if SUDARSAN wrongly flags content or accounts. That leaves a key distinction, and a key question, around the current system. Unlike the 2018 proposal, SEBI has not said SUDARSAN monitors private messages or emails. But the regulator is now using AI to continuously scan, analyse and classify large volumes of public online speech. The system looks for possible violations. This brings questions around transparency, accuracy and oversight back into focus. It also raises questions about the limits of automated regulatory surveillance.