Why human-AI skills matter in financial decision-making

New research suggests the value of AI in retail investing depends not only on the technology itself, but on whether investors can understand and use AI-generated advice responsibly. The authors, M. Vijayananth and N. Saravanabhavan, claim that AI trading tools can build investor trust, encourage sustainable investment behavior and support financial resilience, particularly when users have stronger human-AI interaction skills.

Titled Beyond algorithmic trust: human–AI interaction competency strengthens AI-driven financial decision-making and financial resilience and published in Frontiers in Artificial Intelligence, the study analyses how AI trading usage and financial uncertainty shape the behavior of middle-income retail investors in India. The researchers found that trust in algorithms plays a major role in connecting AI use to sustainable investment decisions, while human-AI interaction competency determines how strongly that trust turns into responsible financial behavior.

AI trading is changing how retail investors make decisions

AI-based trading systems, robo-advisory tools and predictive analytics can process large volumes of data, detect market signals and generate investment recommendations at a speed individual investors cannot match on their own. In fast-moving markets, AI tools promise faster analysis, more structured decision support and wider access to market intelligence. However, the study warns that access alone does not guarantee better decisions.

The research focuses on middle-income retail investors in major Indian cities, a group that is increasingly active in equity markets and exposed to volatility in an emerging financial environment. The study surveyed 569 investors from Mumbai, Chennai, Kolkata, Delhi, Bengaluru and Hyderabad between January and March 2026.

The authors used a cognitive-behavioral framework known as the Stimulus-Organism-Response model. In this model, AI trading usage and perceived financial uncertainty act as outside triggers while perceived algorithmic trust is the internal cognitive response. Sustainable investment behavior and financial resilience are the outcomes.

The framework reflects a key concern in AI-driven finance: investors do not respond to technology mechanically. They first decide whether they trust the system, which then shapes whether they use AI insights in disciplined, long-term and sustainable ways.

Algorithmic trust links AI use to sustainable investing

AI trading usage was found to have a positive effect on perceived algorithmic trust. Investors who used AI tools more often were more likely to trust algorithmic recommendations, view them as reliable and feel confident applying AI-generated insights to trading decisions.

Financial uncertainty also increased algorithmic trust. When markets felt unstable or difficult to predict, investors were more likely to rely on AI systems to interpret signals and reduce ambiguity, suggesting that uncertainty can push investors toward algorithmic decision support, especially when they believe the technology offers stability and analytical depth. This trust influenced sustainable investment behavior. Investors who trusted AI-based recommendations were more likely to consider long-term returns, responsible companies and environmental, social and governance-related factors in their investment choices.

Sustainable investment behavior is more than an ethical preference. It is treated as a disciplined financial approach that can help investors avoid impulsive decisions and remain focused during uncertain market conditions. The research suggests AI tools can support responsible investing when they help users process information more clearly and act with a longer-term view. However, the path from AI use to better outcomes is indirect. AI trading does not automatically create financial resilience; it works through trust and behavior.

The strongest direct link in the model was found between sustainable investment behavior and investor financial resilience. Investors who followed more sustainable and disciplined investment strategies were more likely to report the ability to handle shocks, recover from losses and remain financially stable. The study also found that perceived algorithmic trust and sustainable investment behavior worked together as a sequential pathway. AI use and financial uncertainty first shaped trust in algorithms. This trust then supported sustainable investment behavior, which in turn strengthened financial resilience.

Human-AI competency makes the difference

In addition to trust, investors also need the ability to understand, interpret and evaluate AI-generated advice. Human-AI interaction competency strengthened the relationship between algorithmic trust and sustainable investment behavior. Investors with higher competency were better able to convert trust in AI systems into responsible investment choices. They were more likely to understand algorithmic outputs, assess whether recommendations were useful and apply AI insights in a measured way.

The study suggests trust can be productive only when paired with user competence. Without that competence, investors may misunderstand recommendations, over-rely on automated outputs or apply AI advice without enough judgment. The research thus draws a line between using AI and working effectively with AI. An investor can trust a trading system, but that trust becomes more valuable when the investor can question the recommendation, understand its limits and connect it to a broader investment strategy.

AI tools that are fast and accurate may still fall short if users cannot understand how recommendations are produced or how risks should be weighed. Clearer explanations, stronger interface design and better investor education could increase the value of AI-driven systems.

The study also highlights the risk of blind reliance. Algorithmic recommendations can appear authoritative, especially during uncertainty. If investors trust them without understanding them, AI could encourage poor decisions rather than resilience. The authors argue that stronger human-AI interaction skills can reduce that risk by helping investors evaluate outputs instead of simply accepting them.

For India's expanding retail investor base, this is crucial because many investors are entering markets with varying levels of financial and digital literacy. AI may widen access to investment intelligence, but uneven user competency could also widen gaps in decision quality.

Implications and limitations for AI-driven finance

The study has implications for investors, financial platforms, technology developers and regulators.

• Investors should use AI trading tools as decision-support systems, not substitutes for financial judgment. AI literacy and financial literacy are becoming linked skills in modern investing.
• Financial service providers must focus on transparent and user-friendly AI systems. Platforms should explain the basis of AI recommendations, show risks clearly and encourage long-term decision-making rather than speculative trading. Responsible design could help investors use AI tools in ways that support resilience rather than impulsive behavior.
• For AI technology providers, the research suggests that technical performance is only one part of adoption. Tools must be understandable, usable and reliable. If investors cannot interpret AI outputs, even trusted systems may fail to improve decision-making.

The study also raises some policy questions. Regulators may need to strengthen rules around explainable AI, investor protection and algorithmic transparency in trading platforms. Risk disclaimers, behavioral safeguards and digital literacy initiatives could become more important as AI becomes embedded in financial markets.

A balanced approach is what the authors advocate for. AI-driven trading can support better financial decisions, but its benefits depend on responsible governance and investor capability. Transparent systems, strong oversight and user education are central to preventing over-reliance and helping investors apply AI insights effectively.

Notably, the research has limitations. It used a cross-sectional survey, which means it cannot prove causality over time. The findings are based on self-reported responses, which may not fully match actual trading behavior. The study also focused on middle-income retail investors in India, therefore, the results may not apply equally to other investor groups, income levels or countries.

The authors hope that future research could refine the framework, use actual trading data, test different investor segments and examine how trust in AI changes across market conditions.