THE world of finance is an ideal testing ground for exploring the possibilities of artificial intelligence (AI). We’re already seeing how data-driven technologies help influence our financial behavior: there are tools that use AI to assist users in investment research, and there are local apps that help people track expenses better, and that even recommend credit cards that best suit consumers’ needs.
The Bangko Sentral ng Pilipinas recognizes the convenience and operational efficiency that AI can bring to the table. Machine Learning, a subset of AI that learns patterns from data to help formulate decisions, is seen by the BSP as a tool that could enhance credit scoring, client interfacing, and even insurance risk management activities.
Role of AI in credit scoring
AT TransUnion, we have been using explainable machine learning for quite some time. We couple this with historical big data that we’ve aggregated. We analyze this data, identify patterns from it, and use the insights derived from it to create more personalized and fairer assessments of an individual’s credit profile.
On a larger scale, we believe that AI can also help cast a wider net towards greater financial inclusion. Through its capabilities to process large amounts of data, AI-enhanced credit scoring can extend opportunities to unserved or underserved Filipinos. By considering and analyzing data gathered from alternative sources in the credit scoring process, those who are new to credit and those with thin-file credit history stand to reap the benefits.
WHILE much has been said about the opportunities that come from the financial sector’s use of AI, unique challenges may also arise. Data quality is a crucial consideration for the implementation of AI in financial services. With the accuracy and effectiveness of AI models dependent on the quality of data used to train them, poor data quality can lead to inaccuracies in results and predictions.
Certain ethical considerations surrounding fairness must be addressed as well. Without any form of human oversight, the quality of data fed to AI algorithms could perpetuate existing biases towards or against individuals or groups. This could significantly impact outcomes in terms of loan approvals, investment decisions, or insurance underwriting.
Additionally, cybersecurity concerns can also crop up at different stages of AI implementation. Malicious actors can target a financial institution’s data and manipulate it —compromising integrity. Unauthorized data access can also lead to vast reserves of data being stolen and used to target individuals or organizations.
Why AI needs us
CERTAINLY, the opportunities for AI to transform the financial industry are massive. However, while it’s easy to be amazed by AI and its power, I can’t help but think of the advice a man named Ben Parker gave his nephew Peter (Spider Man): “With great power, comes great responsibility.”
As AI technology continues to evolve, the measures we, in the formal financial sector use to keep data safe must change as well. To prevent the biases that exclude Filipinos from the life-changing potential of having access to credit, we must establish stringent data governance practices and continuously monitor data quality. As AI systems are trained on historical data, we must take it upon ourselves to incorporate our expertise as markets and customer behaviors can change at the drop of a hat.
Above all, as efficient as AI can make things for us, we must never forget that it is only technology, and it will always be up to us to determine just how well we use it to make lives better.
Pia Arellano is the president and CEO of Transunion Information Solutions Inc. (TransUnion PHL). She has over 28 years of industry experience across banking, payment solutions, telecommunications, and remittance services. The views and opinions expressed herein are those of the author and do not necessarily represent the BusinessMirror. E-mail questions to email@example.com.