Banking GenAI #7 - Insights for COOs
Welcome to the latest edition of GenAI for Banking COOs. We hope your year is off to an exciting start and we are eager to share the latest and greatest across the world of GenAI in Banking. In today’s newsletter, we cover:
**********
1. GenAI Case Studies from Around the Globe – Triumphs and tribulations
In this insightful article, Liz Lumley explores the fast-evolving landscape of banking, and the reality that the GenAI revolution is not a distant concept but a reality reshaping the industry today. With 87% of financial firms either using, planning to use, or evaluating GenAI, the wave of change is undeniable. The article highlights NatWest, for instance, which has enhanced its virtual assistant, Cora, through GenAI implementation. Cora+, powered by watsonx, offers a leap in customer interaction, providing natural, conversational responses and accesses many other secure data sources compared to what was available in the chatbot earlier.
However, as financial institutions embrace this transformative technology, they must tread cautiously to avoid potential pitfalls like discrimination and inaccurate data. NatWest's early experimentation, involving an AI misidentifying a mortgage applicant as 'Monica' from Friends, underscores the need for vigilance in managing datasets. An internal test case for a mortgage application by ‘Monica’ that was designed to succeed got rejected because the LLM remembered that ‘Monica’ had been fired in a Friends episode. We must confess this example really made us sit back and think. Thank you Natwest and Liz Lumley for bringing this to us.
The article goes on to provide many other examples in detail. Neobank bunq's Finn, a GenAI platform, redefines user-centric banking by understanding context and answering advanced queries. Fintech company Bud Financial introduced Drive, an AI-powered data intelligence solution with a GenAI chatbot, Drive Copilot, empowering employees to identify opportunities. We strongly encourage you to read the article in detail for many insights.
Net net, while the financial industry speeds toward innovation, the cautionary tale of NatWest’s ‘Monica’ experiment reminds us of the importance of data integrity and transparency in ensuring positive client outcomes amid the GenAI revolution.
?**********
2. GenAI use cases: A three step ladder
In this article in the Harvard Business Review, Scott Cook, Andrei, Haigu, and Julian Wright lay out an interesting three step ladder concept to thinking about GenAI use cases. Level 1 is all about accessing knowledge and insights from easily available data sources. Think of internal knowledge access use cases or answering basic questions via a client-facing chatbot. Level 2 is about creating a more domain-specific customized solution that brings in insights developed from the business that the company is in. In banking context, think about Generative AI private bankers. Level 3, the final step, is where you continuously refine the solution through a data feedback loop. Think about a continuously improving Gen AI private banker that is learning from ongoing real-world feedback from customers. This leads to some interesting design choices on how such feedback should be collected from customers, but the underlying concept is sound. ?Of course, the implementation complexity also increases as you go up the ladder.
As banks accelerate use cases adoption, we figured the above framework might be a helpful way to think about where these use-cases stack, and how fast are they moving up the maturity ladder in their implementation.
**********
3. Yes, GenAI is already being used for fraud and it is only accelerating. GenAI is also the solution
AU10TIX has written a fascinating piece on Venturebeat about how GenAI is an identity fraudster’s best friend (and worst enemy). While Generative AI has various use cases to improve banking from customer experience to operations, the same technology can be used to facilitate identity theft, fraud, and swarm attacks. AU10TIX's annual report reveals a 44% surge in organized identity fraud, resulting in significant financial and reputational damage. Bad actors can convincingly mimic identities with techniques like deepfakes, bypass biometric ID verification, and even manufacture entirely new personas all with the help of generative AI.
领英推荐
To counter this threat, organizations are advised to implement a multi-layer defense strategy, integrating generative AI detection tools alongside existing AI and ML detection and verification systems. For COOs, ensuring that the IT Security colleagues are incorporating generative AI into defense systems becomes critical to effectively combat these increasingly sophisticated attacks. Recognizing the vulnerability of your KYC and risk management processes to generative AI-enabled attacks is the first step. COOs must adapt their strategies to leverage the same technology in mitigating risks, ensuring a proactive stance against emerging threats.
**********
4.?$1.5bn and counting – AI-delivered value at JPMC: what can it do for your bank?
Financial Institutions have been cautious in deploying AI tools for client-facing purposes. Quite a few do not expect to see it in Production until 2025 though work is going on in earnest. However, there is a noticeable move towards leveraging AI for back- and middle-office projects. As Aaron Ricadela from Oracle highlights, several banking executives in the Evident AI Symposium are using AI to train pattern-spotting and process-automating AI software to manage back-office functions. The allure lies in the realization of cost savings and opportunities for revenue generation. As highlighted in the article, JP Morgan’s Theresa Heitsenrether, who is heading their global AI efforts, mentioned that the bank is expected to realize more than $1.5bn in value this past year from running AI tools to streamline operations, detect fraud, screen transactions for potential sanctions violations, and render credit decisions.
While the broader risk and regulatory framework evolves, the value creation seen at JPMC underscores what AI (and GenAI) can do for banks. We worry that the medium-sized banks may not have the luxury to wait and see how the field is evolving. When it comes to operational activities, the big banks will continue to bring their marginal cost significantly down, further widening the gap to the smaller banks. If you are a COO in a mid-sized bank, this puts you in a tough spot.? Not doing anything is not an option. We encourage making smart bets and getting started on this journey to build the organizational muscle to scale it fast as the field evolves.
**********
5. Accenture’s Top 10 banking trends for 2024: How to Reconnect with Customers through AI-Powered Conversations
Each year, Mike Abbott, the global banking lead at Accenture share his views on the trends that are most likely to drive impacts in the banking industry in the coming year and beyond. One of the key trends called ‘Capturing the Digital Dividend’ is the need for banks to focus on finding way to drive deeper and more meaningful engagement with their customers across digital channels. As customers have moved to digital channels, customer relationships have suffered, leading to weakened loyalty and brand differentiation. Digital channels prioritize service over meaningful conversations about financial aspirations and goals, limiting the potential for personalized advice and proactive support. This is the problem that GenAI can help banks tackle and is the focus of intense activity in 2024. We are curious to see what comes out of it, because as customers, all of us will benefit from bank’s cracking this tough problem. Good luck to all the teams. We are rooting for you.
**********
This edition just scratched the surface of GenAI's potential in the coming year. Buckle up because the next chapter of banking is unfolding fast. Stay tuned for further insights, industry perspectives, and practical guidance as we chart the course towards an AI-powered future. See you on the road ahead!
#GenAI #COO #BankingOps #Innovation #ArtificialIntelligence #BankingTechnology #Banking #CapitalMarkets #ZeroOps #Transformation
?
This issue of the newsletter is brought to you by Accenture’s Rajat Dev | Nazat Dowla, MBA / MPP | Ashley Iu | George Heathcote | Ian Jacobs | Donrick Sanderson II | Sebastian Cadillo