The Ethical Imperative of AI in Financial Services

As artificial intelligence (AI) becomes central to financial services, the stakes have never been higher. With billions of dollars pouring into AI-driven tools for credit scoring, fraud detection, and product innovation, industry leaders face a critical question: How can we ensure that financial intelligence remains ethical, transparent, and inclusive?

The Expanding Role of AI in Finance

Sneha Walke, founder of The Little Things Co., sparked the conversation by highlighting a stark contrast in media narratives. While one story detailed a fraudulent crowdfunding scheme exploiting victims of an AI mishap, another celebrated AI as a burgeoning wealth generator. With projections indicating that global investments in AI in financial services could reach $97 billion by 2027, Walke posed a tough question: “Where do we start with responsible AI?”

A Global Perspective on AI Governance

Sandhya Vasudevan, a board member of Tie Bengaluru and former MD at Deutsche Bank, offered insights from various global models. She pointed out that while India lacks a unified AI framework, it possesses clear guidelines. “The European approach to risk management is compelling, while China pragmatically considers the impact of AI on economic outcomes. Singapore’s framework emphasizes responsibility and transparency,” she explained.

Data Quality and Governance in AI

Vasudevan emphasized that the majority of AI applications in finance fall into three categories: risk management, product innovation, and customer engagement. For AI to function responsibly, governance and data quality must be prioritized. “Most data we use is historical. In India, access to banking has dramatically improved since 2010. If our AI is trained on biased or incomplete records, we risk excluding the very populations we aim to serve,” she cautioned.

The Necessity of Transparency

Public policy expert Professor Sridhar Pabbisetty introduced the need for transparency in AI systems. He argued that algorithms making credit decisions must be explainable and auditable. “A credit decision must not stem from an opaque algorithm; it should be justified in a manner that can be scrutinized,” he warned, emphasizing the need for clearer communication with stakeholders.

Addressing Bias and Promoting Inclusion

Walke voiced concerns regarding the potential for biased AI models to deny access to marginalized users. Vasudevan countered with realism, stating it could take 10 to 15 years to mend the data quality and representation issues. Nonetheless, she highlighted the opportunity for AI to provide tailored advice that empowers users. “Instead of simply rejecting applicants, AI can outline actionable steps to improve eligibility for credit,” she noted.

Digital Guardianship: A New Paradigm

Vasudevan presented the concept of “Digital Guardianship,” a framework allowing users who are uncomfortable with digital platforms to appoint trusted individuals to manage their finances, complete with clear audit trails. “This is crucial for older citizens and those lacking digital literacy; we must integrate trust mechanisms into our financial systems,” she advocated.

The Regulatory Role in AI Development

Lastly, Walke sought reflections on the regulatory landscape surrounding AI. Pabbisetty underscored the need for regulators to look beyond mere compliance. “India’s fintech environment is shifting toward more cost-based advisory services, free from commission incentives. Regulators should support these initiatives and actively pursue predatory practices,” he suggested, also highlighting the need for improved fraud awareness and rapid recovery mechanisms.

This roundtable discussion at the BW Businessworld Festival of Fintech 2025 culminated in a shared sense of urgency. As AI’s role in financial services expands, it is essential to develop systems that are not only technologically advanced but also ethically grounded.