Reserve Bank of India: Rapid AI adoption pose financial stability risks

Getting your Trinity Audio player ready...

The Reserve Bank of India (RBI) has cautioned that the growing popularity and adoption of artificial intelligence (AI) by the financial sector come with several risks despite the benefits. 

In its December 2024 Financial Stability Report, the central bank highlighted the significant market concentration risk within the financial industry and key third-party providers of cloud and AI services.

“The evolution and adoption of AI poses several risks to financial stability,” RBI pointed out in the report. 

“First, interconnectedness could become enhanced through over-reliance on shared technology, service providers and infrastructure. In particular, there is a high risk of market concentration both within the financial industry as well as critical third party service providers of cloud and AI services,” the report stated.

Second, the risk of cyber threats escalating into a financial stability issue is high, as AI could empower cyber attackers with advanced phishing techniques, including creating deepfakes using generative AI. According to the RBI report, the increasing availability of AI services like ChatGPT has raised concerns about their potential use in cyberattacks.

“Third, according to the International Monetary Fund (IMF), the increased adoption of AI in capital markets can create additional risks related to increased market speed and volatility under stress, especially when trading strategies using AI become highly correlated. Specifically, if such trades are funded through leverage, any shock could amplify market stress through fire sales and feedback loops,” RBI said in the report.

Moreover, AI may encourage migration of more activities to non-banking financial institutions (NBFIs), increasing systemic opacity.

Fourth, if technological penetration and market and vendor concentration are high, risk transition from individual firms to the financial system could be nonlinear and portend systemic risk, RBI’s report stated. 

The central bank’s warning comes at a time when generative AI is expected to contribute $359-438 billion to India’s gross domestic product (GDP) by 2029-2030, making ethical and responsible use of the technology imperative for the country’s economic growth and financial sector. 

In 2024, Bank of Baroda, one of India’s leading public sector banks, introduced a generative AI-powered virtual relationship manager to enhance the digital customer service experience. A relationship manager offers advice on investments and available financial services.

Last year, the RBI also acknowledged an increased mention of AI technologies in the annual reports of public sector banks, a shift from the previous trend where private sector banks were more proactive in adopting AI. In its October bulletin, the central bank highlighted that the enthusiasm for AI technologies in many public-sector banks now matches that of their private-sector counterparts.

‘Model risk’

Financial institutions are also prone to increased risks in terms of bias and hallucination, misuse, over-reliance on common models, faulty predictions, data quality issues, and third-party dependence.

Model risk would be a major hazard for financial firms that use AI tools, RBI said in its report.

“A key challenge with AI models is their lack of explainability or the so-called ‘black box problem’ due to the difficulty in explaining how these complex models are making decisions even as they achieve more accurate predictions,” RBI said.

Since AI relies heavily on the data it is trained on, the lack of transparency in how these systems function could lead to models using biased or less related data. These concerns are especially critical in the financial sector, particularly within the rapidly growing banking industry.

Enjoy the benefits and tackle the hazards

In its Financial Stability Report, the RBI recognized that generative AI models would be transformative due to their unique ability to adapt and learn autonomously at speed, generate a wide range of responses in various formats rather than being limited to predefined options, and, in certain cases, match or surpass human capabilities.

According to the RBI report, the ever-evolving AI technology gives financial firms advantages such as the ability to process vast amounts of unstructured data, scalability, and adaptability. These features could lead to efficiency gains and cost savings across various areas, including quantitative analysis, risk management, operations, customer interactions, and cybersecurity.

However, RBI’s report pointed out that while rapid technological advancements and adopting new technologies bring benefits, they also pose risks to the financial system. As a result, financial sector regulators and supervisors are intensifying their efforts to enhance regulation and oversight to mitigate potential financial stability risks from emerging technologies.

In December 2024, the RBI formed a committee to develop a Framework for Responsible and Ethical Enablement of AI (FREE-AI) in the financial sector, bringing together experts from various fields. The Committee will evaluate the current adoption of AI in financial services, both globally and in India, identify potential risks, and recommend frameworks for evaluation, mitigation, and monitoring for financial institutions.

In the securities market, the Securities and Exchange Board of India (SEBI) has directed market infrastructure institutions and registered intermediaries using AI tools—whether developed in-house or sourced from third-party technology providers—to assume full responsibility for their use. They must ensure the privacy, security, and integrity of investors’ data, regardless of the scale of adoption.

“Standard setting bodies and national regulators and supervisors should, therefore, take a balanced approach to reap the benefits of AI while safeguarding the financial system. They must update their skills and tools as well as proactively adapt their frameworks to identify and mitigate emerging risks from this rapidly evolving technology,” RBI’s report added.

In order for artificial intelligence (AI) to work right within the law and thrive in the face of growing challenges, it needs to integrate an enterprise blockchain system that ensures data input quality and ownership—allowing it to keep data safe while also guaranteeing the immutability of data. Check out CoinGeek’s coverage on this emerging tech to learn more why Enterprise blockchain will be the backbone of AI.

Watch: Blockchain, IPv6, AI & 5G will pave the way for the new Internet