Five priorities for harnessing the power of GenAI in banking

Banks face significant challenges that may constrain their ability to generate strong returns from their investment in GenAI. The survey results identified a few main challenges:

Insufficient expertise and resources

Banks see gaps in resources for staffing GenAI initiatives. More than half of survey respondents said insufficient internal expertise was a top challenge in establishing a dedicated GenAI team. 

Cost and budget constraints

Economic realities are limiting banks’ investments in all technologies and GenAI is no exception. More than half of survey respondents cited implementation costs as a challenge when exploring GenAI initiatives. 

Lack of confidence in internal capabilities

The aged, heavily-customized technology architectures in place at many banks today, with all their workarounds and poor data flows, are a barrier to AI implementation. Recognizing these constraints, a significant proportion of survey respondents said they did not believe their institution had the correct technological infrastructure and capabilities to implement GenAI.

Prioritizing use cases

The competing options for deploying AI challenge banks to identify the most impactful initial use cases. Many banks are prioritizing legacy automation capabilities (e.g., robotic process automation) in back-office functions. A clear majority of respondents say their banks are waiting for further development and testing before prioritizing front-office use cases.  

Regulatory uncertainty and risk

Evolving regulations create uncertainty about compliance requirements and the liability risks banks could face. From a resiliency perspective, banks need to be prepared for hackers, fraudsters and other bad actors taking advantage of the power of GenAI. Because regulation is catching up, firms will need to think about how they build and enable systems that anticipate developments in regulation, rather than building processes that might be overtaken by restrictions. Similarly, banks looking to deploy must bear in mind regulators’ claims that existing rules will apply to GenAI.

Risks related to data privacy, security, accuracy and reliability are banks’ top concerns for GenAI implementations. That’s understandable given that large language models (LLMs) can be subject to hallucination and bias. The prevalence of sensitive and confidential data in banking raises concerns about accidental data breaches and erroneous transactions. 

2. Explore an ecosystem approach to access new technology and talent

The many banks that need to update their technology could take the opportunity to leapfrog current architectural constraints by adopting GenAI. However, for GenAI to be useful in the workplace, it needs to access the employee’s operational expertise and industry knowledge.

Given the newness of GenAI and the limited tech capabilities of many banks, acquisitions or partnerships may be necessary to access the necessary skills and resources. GenAI’s ability to work with unstructured data makes it easier to connect and share data with third parties via ecosystems. Half (51%) of banks said they prefer partnerships as their go-to-market approach for GenAI use cases, as opposed to in-house development.

Where to act now

Identifying opportunities to modernize infrastructure, enhance data quality and improve data flows is the critical first step. 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.

Looking ahead

Acquisitions and joint venture opportunities can help banks build new or enhance existing GenAI-focused ecosystems and deliver new products and solutions more quickly. The business case for such deals should be based on a careful assessment of capabilities and with results from initial use cases. 

3. Rebalance the innovation portfolio while identifying use cases

When it comes to GenAI specifically, banks should not limit their vision to automation, process improvement and cost control, though these make sense as priorities for initial deployments. GenAI can impact customer-facing and revenue operations in ways current AI implementations often do not. For example, GenAI has the potential to support the hyper-personalization of offerings, which helps drive customer satisfaction and retention, and higher levels of confidence.

Similarly, many banks have been pursuing industry verticalization and deposit retention strategies, as well as seeking new and diversified revenue streams. These are logical topics for a discussion of initial GenAI use cases.

5.  Establish governance and controls

GenAI introduces new and heightens existing risks in banking operations. 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.

Organizations must consider when and how employees can leverage GenAI and evaluate the distinct risks of internal and external use cases. GenAI’s impact on operations is another factor. For example, the application of GenAI to lending decisions could lead to biased outcomes based on protected characteristics (e.g., gender or race). The burden of proof rests with banks, meaning they will need to collect evidence to show regulators why applications are denied and that applicants are considered fairly. Even where there are no legal or regulatory boundaries at present, governance models must be designed to promote responsible and ethical use of GenAI.

Where to act now

As a first step, banks should establish guidelines and controls around employee usage of existing, publicly available GenAI tools and models. Those guidelines can be designed to monitor and prevent employees from loading proprietary company information into these models. Additionally, top-of-the-house governance and control frameworks must be established for GenAI development, usage, monitoring and risk management agnostic of individual use cases.

Looking ahead

As banks make further investments in GenAI capabilities and develop new use cases, they’ll need to assess the unique challenges and risks associated with the tooling and adjust governance and controls for each individual use case. New use cases will bring new ongoing requirements for testing and assessments for hallucination, bias and other risks.

In closing

Taking advantage of the transformational power of GenAI requires a combination of new thinking about a longstanding challenge for banks — how to innovate while keeping the lights on. But banks clearly understand the urgency; a huge majority are already dedicating resources to GenAI.

Success in GenAI requires future-back planning to set the vision and a programmatic approach to use-case prioritization, risk management and governance. Banks will need to challenge their current understanding of AI primarily as a technology for back-office automation and cost reduction. Thinking through how GenAI can transform front-office functions and the overall business model is essential to maximizing technology’s return on investment. Banks must also recognize GenAI as just one piece of an overall innovation agenda. Using GenAI along with a balanced set of measured actions supported by a longer-term strategy will allow banks to create value for customers and shareholders while building the bank of the future.

Summary

To harness the true power of GenAI, banks will need to assess use case value and risk to inform a longer-term roadmap. Reviewing lessons learned from technology innovation projects, data management capabilities and talent, banks can help develop a framework for use case development. Establishing enterprise governance and controls for internal and external GenAI usage and a control tower approach will be critical for assessing use case value creation while also managing associated levels of risk.

Prashant Kher,  Senior Director Digital Assets Strategy Lead, EY-Parthenon, Ernst & Young LLP and Zachary Trull, Financial Services Strategy EY-Parthenon, Ernst & Young LLP  were contributing authors for this article.