I’ve often thought about how AI models can analyze large volumes of financial data, adapt to market shifts, and detect fraud faster than any human could. This kind of problem-solving, especially when it deals with real money and risk, fascinates me. In the finance field especially, AI should be handled with care more than anything. In this industry bias in models can directly affect people’s access to credit, savings, or investments.

On my Padlet wall, I’ll explore three main AI applications in finance: fraud detection, algorithmic trading, and AI-driven credit scoring. I’ll share examples of companies using these technologies, the tools they use (like machine learning or natural language processing), and carry out a risk-benefit analysis.

To me, AI in finance is about building systems that are fair, secure, and smart. I hope to one day work on creating those kinds of systems that help people make better financial decisions.

Banks and other fin-tech companies use ML models to detect unusual patterns in real time to flag suspicious activities. These systems improve with time by learning from both fraudulent and legitimate behavior (Flinders et al., 2025).

Technologies Used are: Machine Learning, Pattern Recognition and collecting and indexing large amount of data.

Benefits: As the title suggests, AI helps prevents frauds and scams happening across people’s bank accounts. It also allows for better auditing for tax purposes.

Example: Mastercard and Visa use AI to scan billions of transactions daily for potential fraud.

Challenges: Everything comes with its own challenges. Since Machine Learning models are essentially statistical models, there are chances that an output is a false positive, i.e., it might flag an activity as suspicious while it isn’t, causing unneeded human audits and wasting resources.

Personal Reflection: AI’s role in fraud detection interests me because of its ability to learn from its mistakes. Unlike traditional rule-based systems, AI can learn from new types of fraud in real time. It’s a way to stay one step ahead of attackers, which is both technically challenging and socially impactful.

Technology Used: Predictive Analytics, Reinforcement Learning, Natural Language Processing (NLP).

Benefits of AI in Algorithmic Trading: AI brings speed and efficiency as it can analyze vast data sets and execute trades in milliseconds. It can also detect patterns and strategies that humans might miss, improving trading strategies. It also brings in 24/7 trading for certain assets like cryptocurrencies.

Example: Renaissance Technologies and Citadel are leaders in AI-driven hedge funds.

Challenges: Lack of transparency. As some AIs are essentially black boxes, it makes its decisions very hard to explain.

Personal Reflection: I'm drawn to AI-driven trading because it blends finance, data, and speed. It’s fascinating how machine learning models can detect patterns, analyze news, and execute trades in milliseconds, something no human could do. It makes markets more efficient and opens the door to creative, data-backed strategies.

Technology Used: Logistic Regression, Decision Trees, XGBoost, NLP (for alternate credit data).

Benefits using AI: Since AI systems are really fast unlike traditional, human based systems, a customer can get their decisions within seconds. This can seed up load approvals and increase customer satisfaction. It is also very cost efficient when operated at scale while the system can operate with an improved accuracy as the AI has access to large and complex datasets.

Example: Upstart uses AI to assess loan eligibility beyond traditional FICO scores.

Challenges: If the AI is trained on data that has societal or historical bias, its decisions would reflect the same, leading to unfair outcomes. It is also not very transparent as AIs are black-box systems whose results can be sometimes very hard to explain (McKinsey & Company, 2024).

Personal Reflection: I find this area meaningful because AI has the potential to make lending more inclusive. Traditional credit systems often miss people with little formal financial history. AI can look at broader data, like spending habits or alternative income, to assess risk more fairly and help more people access credit responsibly. However, this must be handled with care as AIs can also have unintentional bias. Excluding groups of people is the opposite of what this technology aims to accomplish.

In addition, predictive analytics will also become even more sophisticated, enabling real-time fraud detection and credit/risk assessments to customers (Otieno, 2022). Furthermore, AI will drive the automation of back-office operations such as compliance and regulatory and will increase efficiency and compliance with local law.

However, these advancements come with significant ethical risks and considerations. The primary concern is algorithmic bias, where AI models that are trained on historical data may perpetuate existing societal biases, leading to discriminatory outcomes in lending, credit scoring, etc. Subsequently, the lack of transparency in AI models makes it almost impossible to understand why a particular decision was made, reducing accountability and trust (Kosinski, 2024).

Data privacy and security are also very important, as AI systems require access to highly sensitive personal and financial information, making them very high value target for hackers, hence raising questions about potential misuse. Job displacement due to automation is another ethical challenge. It has the potential to replace a lot of back-office operations such as regulatory and compliance as mentioned above, affecting the lives of hundreds and thousands of people. These considerations must be taken seriously before deploying AI in a large, industry-wide scale.

From my perspective, the most pressing ethical concerns revolve around equity, fairness and accountability. It is crucial to ensure that AI systems are developed and deployed with explicit safeguards against bias, and that their decision-making processes are transparent, auditable, and explainable. In financial services, the decisions made by AI systems must be bias free and not exclusionary.

Regarding privacy, the extensive data collection and analysis by AI systems raise important questions and ethical dilemmas about individual autonomy and surveillance. Financial institutions will hold increasingly detailed profiles of individuals, and the potential for this data to be misused, breached, or shared without explicit consent is a significant societal risk. Robust regulatory frameworks and technological safeguards should be put in place to protect sensitive financial information. This should happen even before AI operations are scaled up within the industries.

The issue of equity is dependent on algorithmic bias. If AI systems perpetuate or amplify existing inequalities in access to credit, investment opportunities, or insurance, it could widen the wealth gap and create a less equitable society. This is the opposite of what AI systems in the financial industry is targeted to accomplish. It should be ensured that AI benefits all segments of society, not just the rich and/or the privileged.

My perception of AI in finance is heavily influenced by these societal impacts. While the efficiency and innovation AI brings is certainly impressive, I find the potential for increasing inequality and the decreasing individual privacy to be very concerning. I believe the future of finance, shaped by AI, must prioritize societal equity alongside economic growth, instead of focusing on pure profit and corporate greed.

[Rationale: This article provides a comprehensive view on how AI works in real-time fraud detection in banking]

Kosinski, M. (2024, October 29). What is black box artificial intelligence (AI)? IBM. https://www.ibm.com/think/topics/black-box-ai

[Rationale: This article talks about the black box bias that large AI models face. Essentially, it means that the decisions that AI makes is sometimes very hard to explain or find the rationale behind.]

McKinsey & Company. (2024, July 1). Embracing generative AI in credit risk | McKinsey. Www.mckinsey.com. https://www.mckinsey.com/capabilities/risk-and-resilience/our-insights/embracing-generative-ai-in-credit-risk

[Rationale: This article provides a comprehensive view on how AI can be used in measuring credit risk, and the various challenges it poses.]

Musick, N. (2024, December 20). Artificial Intelligence and Its Potential Effects on the Economy and the Federal Budget. Congressional Budget Office. https://www.cbo.gov/publication/61147

[Rationale: This article provides a comprehensive view on how AI effects the economy as a whole.]

Otieno, N. (2022, February 1). The Future Role of AI in Finance. Www.worldfinance.com. https://www.worldfinance.com/markets/the-future-role-of-ai-in-finance

[Rationale: This article talks about the various future trends and predictions of how AI will evolve in the financial field. ]

Powers, J. (2018). How AI trading technology is making stock market investors smarter — and richer. Built In. https://builtin.com/artificial-intelligence/ai-trading-stock-market-tech

[Rationale: This article talks about how AI is being used in algorithmic stock trading. ]