Sanjay Uppal
Artificial Intelligence has made a transformational impact in many industries, especially the financial services industry. For example, banks and lending institutions use AI models to provide customised financial advice and product recommendations, in addition to proactive fraud detection.
Singapore-based finbots.ai is working with organisations such as banks and lenders to provide credit modelling solutions, and it has already won many clients across Asia. finbots.ai, backed by the likes of Accel Partners, is now in expansion mode.
e27 had a chat with finbots.ai’s Founder and CEO Sanjay Uppal, who shares insights into credit modelling and discusses its importance and use cases in the lending sector and the role of AI in the financial services sector in the coming years.
Excerpts:
How does finbots.ai’s credit modelling solution utilise artificial intelligence, and what kind of data inputs does it use to make predictions?
Our credit modelling solution creditX utilises advanced Artificial Intelligence and Ensemble Machine Learning algorithms to analyse vast amounts of data to develop credit risk scorecards, providing lenders with a faster and more efficient way of assessing credit risk.
creditX can connect to any data source using APIs and develop credit scorecards using various forms of data, including historical data, bureau data, open banking data, or alternative data such as social media and invoicing data. Thus, each of our clients can develop its own custom scorecards in a matter of hours and days instead of six-plus months taken using legacy methods and technologies.
Can you provide examples of how finbots.ai has helped financial institutions make better lending decisions through its AI-powered credit modelling solution?
Our clients span banks, non-finance companies, fintech lenders, e-commerce players, digital banks, and credit bureaus, and each has seen tremendous results with creditX.
One of our first clients, an IFC-backed B2B BNPL fintech lender in Africa with over 200,000 MSMEs on its platform, faced challenges in scaling up profitable lending to its base. creditX enabled it to rapidly build high-accuracy credit scorecards and process thousands of credit applications in a day. This helped it increase the number of applications processed daily by 10x while reducing the loss rates by 5 to 10 per cent.
Another client, a digital bank in APAC, saw its loss rates decline by 50 per cent with minimal impact on loan approvals and revenues, thereby enabling it for faster profitable growth in its lending businesses.
What sets finbots.ai apart from other predictive AI startups in the financial industry, and what competitive advantages does it bring to the table?
In the predictive AI space, most solutions are in the form of AI toolkits that have high skill dependency on data scientists, data modellers, analytics specialists and domain experts for the development of high-quality models. The complexity, skill dependency, and required vendor support results in significant time requirement and an overall high cost of ownership. Additionally, organisations often have to expose their confidential data to get the desired models.
Also Read: How Transparently.AI uses Artificial Intelligence to detect accounting manipulation, fraud
finbots.ai has built a vertical AI product creditX which pioneers predictive AI for the credit modelling use case, taking into account all regulatory, data privacy, security and risk management challenges the industry faces.
In comparison, building a model that takes weeks with an AI toolkit takes less than an hour on creditX.
Today, only the largest 5 to 10 per cent of lenders can afford sophisticated credit modelling solutions. We are democratising credit modelling, making creditX affordable for even the smallest lender, with a pricing model that scales as a lender grows. Hence, our clients range from small start-up lenders to mid-to-large-sized banks.
Our credit modelling solution, creditX, allows organisations to perform a rapid Proof-of-Concept (PoC).
How does finbots.ai ensure the accuracy and reliability of its credit modelling algorithms, and how often are they updated?
We have a rigorous process to ensure the accuracy and reliability of our credit modelling algorithms, including extensive testing and quality control measures for a variety of scenarios. We are also constantly evolving our models to incorporate new advances in Data Science.
In fact, finbots.ai is amongst the first companies to complete ‘AI Verify’ – the world’s first AI governance testing framework and toolkit developed by Singapore’s Infocomm Media Development Authority (IMDA) and the Personal Data Protection Commission (PDPC).
The AI Verify framework enables companies to objectively validate the performance of their AI systems, ensuring that they are fair, explainable, and responsible.
What are some of the biggest challenges that you face when working with financial institutions, and how does the company address them?
For financial institutions, the process to develop and deploy a credit model has evolved little over the past three decades and takes six to 12 months. Only a small fraction of lenders today have access to and affordability of the technology and skills required to develop sophisticated credit scorecards.
creditX delivers higher quality credit scorecards with the entire process completed in hours. This requires a significant paradigm shift in credit management and transforms parts of the credit risk management process.
To facilitate this shift, we offer clients access to sandbox environments to test our solution and bring their own data for Proof of Concept (PoC) testing, which can be completed in 3 to 5 days. After all, seeing is believing.
How has finbots’s client base evolved since its inception, and what are its plans for expanding its reach in the future?
We launched version 1.0 of creditX in early 2020, with our first client deployment in Dubai. Soon after that, COVID-19 hit, and we focused our energies to accelerate product development. Our current version was originally envisioned for Q4 2024.
Since completing our Series A funding round in April 2022, we have expanded our client base to institutions across Asia, Africa, the Middle East, and most recently Australia. Our clients span the full breadth of lending organisations from banks, fintech lenders, mortgage lenders, credit bureaus, e-commerce lenders, loan marketplaces, SME lenders, and agri finance lenders.
We continue to deepen our market presence in Asia Pacific, Africa, APAC, the Middle East and India throughout the year and will expand to North America later this year.
How does finbots.ai’s approach data privacy and security, especially given the sensitivity of financial data?
Credit risk is a highly complex area with demands from both regulators and boards for model development, explainability of AI, data security and data privacy.
Our leadership team brings deep global experience in financial services that we have harnessed in our product design that ensures our product meets these stringent requirements of regulators and boards.
Also Read: From human to AI: Embracing change and thriving in the new world of work
finbots.ai is ISO27001 certified and recognised as SOC2 Type 2 compliant, ensuring the highest levels of data security & privacy in our product and operations, thereby giving our clients the required confidence.
creditX is designed to comply with regulatory standards across markets and is evidenced by our presence in 8 markets, and growing more.
How does finbotsAI see the role of AI evolving in the financial industry over the next five to ten years, and how does it plan to stay at the forefront of these developments?
AI will have a transformational impact on financial services across areas of services, customer service, process efficiencies and risk management.
The approach we have taken towards AI is to build solutions that enable rapid time-to-value for clients, and this remains core today.
We have a well-defined roadmap over the next five years, including launching other products that transform areas of financial services that have remained relatively unchanged for the past 30 years.
The company has clients across India, Dubai, Singapore, Australia and Mongolia.
We also have a growing client presence in Southeast Asia and Africa. In addition to these regions, we have also tested the suitability of our product in North America, South America and Europe. We have plans to expand to these markets later in the year.
We believe our AI-powered credit modelling solution can help financial institutions across the globe, and we are committed to bringing it to as many markets as possible to drive financial inclusion and help lenders mitigate risk.
—
Echelon Asia Summit 2023 is bringing together APAC’s leading startups, corporates, policymakers, industry leaders, and investors to Singapore this June 14-15. Learn more and get tickets here.
Echelon also features the TOP100 stage, where startups get the chance to pitch to 5000+ delegates, among other benefits like a chance to connect with investors, visibility through e27 platform, and other prizes. Join TOP100 here.
The post AI will transform customer service, risk management in financial services: finbots.ai CEO appeared first on e27.