News!: Opus Technologies Launches FinGeniusAI Solutions – An Open Innovation Platform for Building Future-Ready Solutions.. Know More
News!: Opus Technologies Launches FinGeniusAI Solutions – An Open Innovation Platform for Building Future-Ready Solutions.. Know More

CEO’s Corner

The Price of Progress: Understanding the Cost Factors Involved in Integrating AI Technology

Share :

CEO’s Corner Series: Unlocking Intelligent Banking: A Deep Dive into the Impact of AI on Banking and Payments

About this series:

In this exclusive series, we will explore the world of Artificial Intelligence (AI) within banking and payments. This series will be a deep dive into how AI is transforming financial experiences, from fraud prevention and hyper-personalization to streamlined transactions and intelligent financial planning.

This is the fifth and final post of the five-part blog series. In previous posts, I have discussed the significant impact of AI in revolutionizing the banking and payments industry. I have discussed various challenges in AI adoption, such as the cost of implementation, identifying the right entry point, navigating a limited talent pool, and achieving faster turnaround times. In this blog post, I aim to explore the most critical question – what is the ROI of an AI investment?

The era of intelligent banking and payments is upon us. Join me and explore the strategies to make AI work for you.

There’s no denying the gold rush mentality around data analytics, machine learning, and artificial intelligence as banks, card networks, payment service providers, and FinTechs allocate large sums of funds to invest in the latest technology. But, as for every prudent business decision, it’s important to measure the return on every dollar spent on AI adoption. Though the AI landscape, particularly the GenAI segment, is relatively new and evolving at a fast pace, we are starting to receive preliminary data to justify the business value of AI investments.

According to a September 2023 study by IDC and Microsoft, companies with mature AI integration are able to realize an average of $3.5 for every dollar spent on AI adoption. It further states that about the top 5% of organizations are able to realize an average of $8 in return. These numbers are across industries and functions; they definitely paint a promising picture for an AI-based future for organizations worldwide.

Source: BCG Analysis, Finding the Sweet Spot for Generative AI in Payments, December 2023

For the banking and payment industry, value creation in terms of operational efficiency and savings is equally important. According to a report by Autonomous Research, AI technologies could reduce operational costs for financial services companies by up to 22%, amounting to savings of $1 trillion by 2030.  The report states that the application of AI in front-office use cases, such as retail banking optimization and reduction in staffing, will lead to $490 billion in savings. Middle-office applications, such as data processing, document analysis, and KYC, will result in $350 billion in savings. Back-office applications, such as underwriting and regulatory compliance, will lead to $200 billion in savings. However, one must remember that even with these estimates, the success of AI hinges only on one factor – adding convenience for the customers.

Understanding the Cost Components

The cost of AI implementation is highly dependent on the data, infrastructure, and regulatory requirements of an industry. Unlike manufacturing or retail industries, the financial industry is governed by strict regulatory norms for data security and customer protection. The cost of regulatory non-compliance could be severely damaging. This is evident from the 2023 incident where the US regulators penalized Wells Fargo, BNP Paribas, Société Générale, and 6 other organizations to the tune of $549 million for their employees’ using personal messaging apps to discuss trade deals. For these reasons, an off-the-shelf AI solution may not always work for the financial services industry.

Some of the key cost components to consider are:

Talent acquisition and training: For the successful adoption of AI, financial organizations need specialized skills of data analysts, data scientists, and machine learning programming experts. Banks and FinTechs could choose to hire for these positions. This team will be responsible for setting up the AI roadmap, identifying key uses, and developing, deploying, and testing AI models.

For AI integration across functions, existing silos must be removed. These efforts would require upskilling existing talent to have a working understanding of AI and machine learning.

Data Acquisition and Preparation: Zillow, a house-flipping startup, turned out to be a massive failure because its AI engine was trained on faulty data. This story signifies the role that data acquisition preparation plays in developing the right AI solution. The efficacy of these solutions relies on the availability of clean, reliable, and unbiased data. It is imperative that financial organizations set up proper data pipelines and make provisions for cleaning and evaluating data.

Cloud Infrastructure: AI applications require robust computational resources. To ensure high availability and scalability, most AI applications are developed and deployed on the cloud. Upgrading existing infrastructure to support cloud-native apps and acquiring high GPU hardware add to the cost of AI development.

Maintenance and Upgrades: Like any other IT system, AI solutions require regular maintenance and upgrades. It’s important to factor in these costs while budgeting for AI implementation.

Regulatory Compliance and Risk Management: Given the nature of an AI application, there might be costs related to regulatory compliance. For example, you may have to buy an additional license to comply with standard regulations and undertake custom development for seamless integration with third-party security apps.

To define a budget and stay competitive, it’s important to critically evaluate your implementation approach based on the above-mentioned cost factors. You must always consider the total cost of ownership (TCO) when making a build vs buy decision for an AI-based solution.

Framework to Measure ROI for an AI Project

Having a good understanding of the total cost of ownership for a project solves only one half of the ROI puzzle. To solve the other half, you need to pit the TCO value against the financial and business value derived from this investment. Earlier in the article, we talked about the cost-saving aspect of AI adoption. Along with that, it is important to consider the market opportunity, value generated due to the innovation novelty, and alignment with larger strategic initiatives. With all of these factors in mind, I propose the following framework for understanding the qualitative value of an AI project.

  • Start by creating a list of potential use cases relevant to your banks or FinTech. It is vital that this list aligns with long-term business goals.
  • Evaluate third-party AI applications for these use cases. Gather valuable data based on existing implementations.
  • Identify net productivity savings by understanding the impact of AI on current processes. Given the maturity status of AI in your organization and the available skill set, discount the net productivity savings number in the right proportion.
  • Also, consider time to market based on access to foundational factors, such as data availability and tech infrastructure.

This framework provides a starting point to evaluate the ROI of an AI project. It’s an elementary method to identify the use case that aligns best with business objectives and the highest ROI potential.

Is AI Investment a One-Time Cost?

According to a research report by S&P Global, the bulk of the banks are involved in “incremental innovation” when it comes to AI. With continuous investment, deployment and testing, these financial organizations will scale their initiatives and eventually move to transformational or large-scale projects. This stands true for all other financial institutions as well. As AI evolves, it will penetrate across the functions and deliver much higher value than what we are estimating right now.

To better grasp the continuum of investment in AI, one can explore the AI maturity framework by PwC. The framework categorizes organizations as Starters, Followers, and Leaders in AI adoption. It evaluates the performance of an organization based on the sophistication of AI algorithms being implemented and the degree of operational use cases.

Source: PWC, AI Adoption in Indian Financial Services and Related Challenges, February 2022

Optimizing the Cost of AI Implementation

Investment in AI solutions is inevitable. However, you can follow the principle of proportion to optimize or right-size the costs. For small or medium-market financial institutions, a platform approach might be the right solution. For a larger organization, developing an in-house team that can work with a niche platform that specializes in the financial sector might deliver better results. To gain a larger market share with innovative AI-based solutions, early and efficient engagement is imperative in order to stay ahead of the competition.

Building a proof-of-concept (POC) is one of the simpler ways to optimize cost and evaluate ROI faster. With an extensive industry experience of 26 years, our team has developed the FinGeniusAI platform, which promotes open collaboration between partners, solution providers, and clients. This platform has been specifically designed to deliver POCs for use cases in the financial services industry. Currently, the platform is being used to develop POCs for AI implementation in document analysis, payment enrichment, payment routing, cash flow forecasting, and more. Get in touch with our team to understand how the FinGeniusAI platform can deliver a POC for your use case.

This concludes our five-part series on “Unlocking Intelligent Banking: A Deep Dive into the Impact of AI on Banking and Payments”. It was a sincere effort to cover all the aspects of AI implementation for banks, PSPs, credit unions, and FinTechs. AI is at a nascent stage, and it will continue to transform the way we pay and get paid. Hope you liked the content and look forward to your comments and suggestions on LinkedIn.

Opus Official

Praveen TM is a highly accomplished global payments leader with over 20 years of experience. A veteran in the payments industry, Praveen has immense wealth of insights to share in payments innovations, emerging technologies, FinTech, and financial services.

Join our mailing list to be the first to know about industry news, Opus updates & upcoming events

    Please read our Privacy Notice to know how we protect personal data