Unleashing potential: Exploring generative AIs role in banking

Unsurprisingly, banks that are best able to quickly deploy GenAI are looking forward to a return on their bottom line, despite concerns over the human impact of the new tech. By leveraging AI, financial institutions can enhance the efficiency and effectiveness of their IT development processes, ensuring that their technology infrastructure remains robust and capable of supporting innovative AI solutions. This modernization is essential for maintaining competitiveness and addressing the dynamic requirements of the financial industry. RAG implementations involve combining LLMs with external data sources to enhance their knowledge and decision-making capabilities. This integration increases the complexity of AI systems, requiring robust governance frameworks to manage data quality, model performance, and compliance. To address transparency, financial institutions must implement explainable AI techniques that provide insights into how AI models arrive at their decisions.

However, issues including data quality and fragmented legacy systems pose challenges. MENA banks are uniquely positioned to leverage AI, thanks to substantial technology investments and robust regulatory frameworks. Yet, they too face data quality and accessibility challenges, which must be addressed to fully harness AI’s potential. Fintechs remain at the forefront of harnessing gen AI and many of their use cases and solutions are impacting financial services.

This is an opportunity to fundamentally rethink the way the business operates. Around the world, KPMG banking and technology professionals have been hard at work helping clients think through the opportunities, risks and implications of genAI. And while there is still a lot to learn, there are three key themes that continue to resonate. To be clear, banks have every reason to be cautious when it comes to AI — generative AI in particular. Large language models and generative AI systems are trained on massive amounts of data, leaving significant room for bias to creep in. Integrating data-driven AI systems increases the risk of data breaches, requiring continuous monitoring and updates to protect sensitive customer information.

Bankers equipped with Gen AI may find that information searches that once consumed hours could now take minutes. When they need to check up on complex regulations, bankers could, via Gen AI, receive cogent summaries — rather than just citations of, or links to, statutes and other raw material. The latest EY report finds that CEOs recognize the potential of AI but are encountering significant challenges in developing AI strategies. In comparison, just over half (52%) describe their knowledge as high, meaning the vast majority (80%) of banks believe they possess an advanced understanding of AI.

For multinational organizations, cultural differences across regional markets can lead to product misunderstandings, which can create additional regulatory challenges. However, GenAI can help mitigate these regulatory risks, by creating marketing materials across geographies that contain the appropriate tone, language, and cultural references, while also supporting consumer understanding of each product, in each locale. Yet it’s important to remember that GenAI is not intended to replace humans. Rather, the technology augments an existing workforce by increasing processing capacity and quality, while freeing people to focus on relationships and customer facing roles, where human emotional intelligence matters.

Download the complete EY-Parthenon survey insights: Generative AI in retail and commercial banking

Banks may need to enhance computing capabilities (e.g., server capacity, data storage and computational power) to deploy AI in bank’s existing tech and data environments. In addition, building “knowledge graphs” from existing institutional expertise will allow GenAI to extract valuable insight. Banking, like other industries, faces obstacles to GenAI successLike many investments, GenAI does involve a degree of risk and uncertainty. Banking leaders’ foremost concerns involve protecting the privacy (74%) and security (71%) of their – and their customers’ – data.

He’s written about how financial services firms can unlock the full value of generative AI, why the FS adoption of cloud computing has been slower than envisioned and lucrative niches for fintechs moving forward. In addition to his global role, David is the co-organizer of Accenture’s FinTech Innovation Lab, a mentorship program bringing together fintech start-ups and leading financial institutions, with labs in the U.K., U.S., and Asia-Pacific. Follow him for continued coverage around how financial services firms and fintechs are embracing technology, AI and data to reinvent their operations and deliver a more personalized customer experience. Gen AI is poised to revolutionize banking by dynamically creating responsive services, potentially adding US$200b to US$400b value by 2030. Existing financial institutions are exploring AI use cases in areas like customer service and risk compliance, demonstrating operational cost reductions.

Where traditional banks have a decisive edge over more recent entrants is the ‘gold mine’ of customer data at their disposal. But outdated core banking systems can often hold back their ability to make the most of this data, leaving the door open to institutions – traditional as well as neobanks and fintechs – with more agile and modern operating platforms. Within the front office, the key advantage is the ability to offer a full range of customers the kind of tailored advice, service and financial solutions that would previously only have been available to high-net-worth (HNW) clients. We’re seeing the results in areas ranging from personalised investment recommendations to virtual assistants that learn from each interaction, engage in a human-like way and relay the resulting data to inform and enrich individual customer insight. Continuous monitoring systems track AI performance and flag potential issues in real-time, while human oversight ensures decisions align with institutional policies and regulatory requirements.

The Future Of AI In Financial Services

“Banks are naturally cautious in embracing generative AI to the full and want to ensure they do so responsibly. A proven track record in delivering responsible AI will be vital for financial institutions to confidently experiment and deploy generative AI models for critical ChatGPT App business functions across the enterprise. Back-up clouds are needed to meet regulatory requirements too, so by leveraging a multi-cloud model, banks and financial institutions are favouring a multi-cloud model where one cloud can act as a backup to another.

These frameworks typically include validation processes, governance structures, and risk mitigation strategies. In some cases, Gen AI technology is useful in identifying the most relevant use cases to pursue. These use cases would be optimized for ROI, ensure integration feasibility, reduce compliance risks or cater to some other set of prioritization criteria. This model ensures critical decisions on funding, new technology, cloud providers and partnerships are made efficiently.

Google & AIR Set Out Framework for AI Risk in Banking Sector

DTTL and each DTTL member firm and related entity is liable only for its own acts and omissions, and not those of each other. This is particularly valuable for financial service organizations, which are not only information intensive, but often have data stored in multiple locations, in the cloud and within local legacy systems. For example, Stanford Digital Economy Lab scholars recently studied3 the impact of a GenAI tool that was deployed at a busy call center. They found that call agents with access to GenAI assistance increased their productivity by almost 14%, with the biggest impact on less experienced workers. In addition, agents with two months of tenure who used the GenAI tool were able to perform as well as agents with six months of tenure who didn’t have that access. The productivity benefits decreased for more experienced employees, which demonstrates that GenAI can make less experienced staff more effective, with, correspondingly, less ramp up time.

Many have predicted the weakening, if not the demise, of Moore’s Law, which says the number of transistors on a computer chip – and therefore its output – doubles every two years at roughly the same cost. And yet Nvidia’s graphics processing units, which power many generative AI models, have increased their computational power by more than a thousand times over the past eight years. Gen AI could streamline know-your-customer compliance and documentation management. Rapidly synthesising client data, it could flag risks and automate paperwork, expediting time-to-ROI.

• This feature improves operational efficiency and reduces manual workloads, allowing teams to focus on more strategic activities.
• This meant customers could contact us via their channel of choice – and instead of queuing to speak to a colleague, they could chat with Cora for help instead.
• In addition, building “knowledge graphs” from existing institutional expertise will allow GenAI to extract valuable insight.
• Nearly one-third (29%) is already using this form of GenAI, and another 33% said they are actively considering it.
• Yet today’s consumers, investors, and corporate customers expect a fast and smooth onboarding experience, plus the best advice and asset management available, quickly.

HKMA’s Arthur Yuen says the Silicon Valley Bank failure holds alarming lessons about digitizing finance. I think that’s a societal pattern we’re starting to see just now – I’ve certainly been getting a lot more voicenotes recently. And if that behaviour is turning up in the way people choose to interact with friends and family, then I think there’s a reasonable chance it will happen with the brands they interact with too. We have to seize the opportunity presented by GenAI to go further, faster and do even better for our customers.

Large language model-powered (LLM- powered) code generation and debugging tools speed up such technology modernization efforts by identifying the underlying business logic constructs. They also help in delivering modern and scalable microservices-based designs. Low-code or no-code Gen AI platforms available in the market automatically generate documentation for the modernized codebase, as well as the API code for integration with the rest of the technology landscape. Some of the more innovative emerging Gen AI solutions also use Retrieval and Augmented Generation (RAG) to iteratively learn and adopt coding standards specific to each financial institution, based on architecture design and standards documentation. Success in GenAI requires future-back planning to set the vision and a programmatic approach to use-case prioritization, risk management and governance.

The adoption of LLMs in financial services is driven by their ability to process and generate human-like text, enhancing operational efficiency and customer experience. Use cases include automating regulatory reporting, analyzing transaction data for fraud detection, generating personalized customer communications, and providing real-time financial advice. LLMs enable financial institutions to streamline processes, reduce operational costs, and deliver enhanced value to customers through advanced analytical capabilities. Financial institutions are prioritizing the integration of AI to address pressing challenges and enhance their competitive edge. Key use cases include automating regulatory reporting, improving fraud detection, personalizing customer service, and optimizing internal processes. By leveraging LLMs, institutions can automate the analysis of complex datasets, generate insights for decision-making, and enhance the accuracy and speed of compliance-related tasks.

Deloitte Global Google Cloud Alliance Financial Services

The IBM Partner Ecosystem is helping banking and financial institutions bring their generative AI dreams to life through IBM watsonx™ Assistant, a next-gen conversational AI solution. The document also addresses the management of third-party AI providers, a crucial consideration as many financial institutions rely on external technology vendors for their AI capabilities. Striking a balance between harnessing its potential and mitigating its risks will be crucial for the adoption of generative AI among financial institutions. EY refers to the global organization, and may refer to one or more, of the member firms of Ernst & Young Global Limited, each of which is a separate legal entity. Ernst & Young Global Limited, a UK company limited by guarantee, does not provide services to clients. We know that before gen AI came along, banks focused on earlier forms of AI, including machine learning, data analytics and regression analysis.

Financial services is one of the many industries where generative AI technology can significantly transform operations. For banks, there’s the potential to tackle challenges such as regulatory hurdles, data governance and rising customer expectations – among others. Customer service, in fact, is another area in which GenAI promises to deliver high impact. As anyone who has ever opened an investment account can attest, new client onboarding involves a lot of filling out and signing of documents, an arduous process for both financial service institutions and their customers. Once a client is on board, there’s still the matter of understanding and managing their assets, and identifying the best opportunities for their particular portfolio – an increasingly challenging task as asset classes expand and become more complex. Yet today’s consumers, investors, and corporate customers expect a fast and smooth onboarding experience, plus the best advice and asset management available, quickly.

Major Gulf banks, including Al Rajhi Bank of Saudi Arabia, Qatar National Bank, and National Bank of Kuwait are already using AI to varying degrees. In the United Arab Emirates, Emirates NBD has partnered with management consultants McKinsey and QuantumBlack—the firm’s AI arm—with the latter reportedly involved in the design and early-stage deployment of ChatGPT generative AI use cases. In mid-August, HKMA launched a GenAI sandbox with the government-funded incubator tech hub Cyberport. The aim is to let financial institutions pilot use-cases within a risk-managed framework and with technical assistance. One of the primary challenges of using generative AI in AML/GFC is the “black box” nature of these models.

These categories would be based on a common substrate that both protects sensitive data and allows results to be compared on a like-for-like basis across multiple LLMs. Indeed, the survey of bank technology leaders indicates that the biggest benefit most banks see from their use of AI and automation is raised employee satisfaction levels. KPMG professionals have talked with employees who are delighted about the increased level of customer service they can provide thanks to automation and AI. Others say they are inspired by working on higher value tasks and activities. Also, while AI can automate and streamline many processes, it should not have the final say in critical decisions such as loan approvals.

While these statistics cover various industries, the banking sector specifically has been heavily reliant on technology since its inception. As the banking industry increasingly moves towards digitisation, the adoption of advanced AI technologies becomes crucial. GenAI, with its ability to synthesise and generate content, offers unparalleled opportunities to automate complex processes, provide personalised customer experiences, and strengthen security measures.

The prevalence of sensitive and confidential data in banking raises concerns about accidental data breaches and erroneous transactions. Enabled by data and technology, our services and solutions provide trust through assurance and help clients transform, grow and operate. Only 6% of those surveyed said their current governance framework is “well-established.” Most (58%) indicated that their frameworks are “in development” – but more than a third viewed theirs as either “ad hoc or informal” (27%) or “non-existent” (9%).

Kundu served as Standard Chartered Bank’s group chief data officer before jumping into the world of AI startups, where he helped promote tools to assist in FIs’ understanding of machine learning. With experience in startups and as a bank CDO, Shameek Kundu explains how genAI is impacting fintech. State-level legislation coming out of Colorado and California may provide more comprehensive guidance, especially as these states deploy GenAI tools for public services. Across the pond, European regulations such as the AI Act are years ahead of early US frameworks and may serve as a helpful guide.

Sidebar is a member-exclusive section where we discuss stories tangential to the main story above. This week’s sidebar is about the recent applications we have seen of Gen AI from the biggest players in the industry. Banks’ strategies should facilitate their employees’ adaptation to a Gen AI future. For some hardcore proponents of AI, the only way to avoid job loss is to upskill, and people ought to have started yesterday.

3 min read – Businesses with truly data-driven organizational mindsets must integrate data intelligence solutions that go beyond conventional analytics. Also based on action.bot from TUATARA and IBM watsonx Assistant, Piotr is a virtual assistant that’s fully integrated with the bank’s knowledge base. In three months, Piotr hit the ground running, taking part in 1,000 conversations over two months. So far, the virtual assistant has achieved a 90% accuracy rate for satisfying support inquiries; a figure that’s expected to rise thanks to built-in learning capabilities. Leon now handles more than 97% of customer conversations without requiring redirection to human agents.

These measures aim to ensure AI systems remain within acceptable risk parameters. Generative AI, a form of artificial intelligence that creates new content based on pattern recognition rather than traditional data analysis, requires specific governance frameworks to manage potential risks, the paper argues. Financial institutions need to revamp their model risk management frameworks to account for the emergence of generative artificial intelligence, according to a new paper from Google Cloud and the Alliance for Innovative Regulation (AIR). We believe that, for too long, banks have had a tradition of investing in tech that results in lower-fee services and unclear customer value.

The traditional approach, relying on manual efforts, not only consumes valuable time and resources but also increases the risk of errors and compliance breaches. “The challenges are deep domain knowledge of treasury management and digital skills gap, complexity of implementation, high cost of digital transformation, increased dependency on legacy technology and organisational siloes,” Hovhannessian concluded. Taking treasury functions as example, Hovhannessian pointed out that Apac banks are increasingly asking for quick deployment, open platform, scalability and resilience from their external fintech partners.

The latest AI Bank Teller utilizes DeepBrain AI’s advanced technology to integrate speech and video synthesis for real-time conversations. You can foun additiona information about ai customer service and artificial intelligence and NLP. A combination of GenAI, NLP and ML represents a paradigm shift in contract management, empowering banks and enterprises to easily manage the complexities of the modern business environment with agility and resilience. By embracing AI-driven solutions, organizations can unlock new opportunities for growth, innovation and sustainable success in an increasingly competitive and rapidly evolving environment. SaaS and cloud banking provider Temenos launches Responsible Generative AI for banking, promising enhanced data management, productivity, and profitability with secure, explainable AI solutions. When banks outsource to third, fourth and fifth parties, some may assume they are also outsourcing risk, but what they may not realise is that by doing this, new risks are created.

One of the biggest and most ubiquitous challenges confronting financial service firms is the matter of rising customer expectations. Today’s consumers demand more personalized experiences, higher quality information, and faster responses. Compounding this, traditional organizations are battling new and more nimble competitors, including robot advisors and digital-first trading platforms, that can meet rising consumer demands and offer results with greater efficiency.

While AI governance processes and controls are somewhat similar to those for legacy technologies, new risks require new models and frameworks, both for internal use cases and use of third-party tools. Strong use cases will include “high-touch” activities historically owned by people, which leverage large datasets or require a generative response logic. Authorities will likely expect firms to deploy advanced GenAI systems in areas like financial crime. Embedded and decentralized finance, tokenization, real-time payments and generative AI (GenAI) are among the powerful forces shaping the banking landscape today. Each presents unique opportunities for banks to reinvent their business models, and GenAI has come to the forefront as a means for banks to accelerate innovation.

Fears are driving some to slow their adoption rather than progressing in a well-thought-out strategy that is flexible enough to be modified without rip and replace. The combined partnership of Oracle and NVIDIA accelerates ‌enterprise-wide adoption of AI. In addition, the concept of the “AI Factory” — systems capable of generating and deploying thousands of AI models across their entire operations representing the next industrial revolution — will be a major theme.

Survey results reflect the latest and most relevant data available from key markets, including the U.S., U.K., Germany, Spain, Italy, Japan, Thailand, Vietnam, Australia, India, Singapore, Brazil, Mexico and China. GenAI is proving instrumental in making digital agents, colloquially gen ai in banking known as chatbots, more personal as well. Today’s GenAI-powered agents are summarizing conversations intelligently, offering similarly conversational responses, acting with human-like empathy, and answering an increasingly complex range of customer requests.

With genAI and a host of other complementary technologies applied, one could theoretically start to run a continuous close. Hook some visualization tools up to that data, and CEOs and decision-makers could tap into a real-time dashboard of key financial, compliance, risk and cost metrics, for example. Generative AI in banking refers to the use of advanced artificial intelligence (AI) to automate tasks, enhance customer service, detect fraud, provide personalized financial advice and improve overall efficiency and security. Our latest 27th Annual CEO Survey indicated that leaders expect technology including GenAI and Machine Learning (ML) to be the centre of optimising costs, creating new revenue streams and improving the customer experience within their organisations. Middle East CEOs are also optimistic about the financial impact of GenAI, with 63% expecting the adoption of it in their organisation to increase revenue, while 62% said it would increase profitability. In the GCC, enthusiasm is even higher with two thirds expecting revenue increases and a similar number expecting profitability increases.