Generative AI (GenAI) is rapidly emerging as a solution to address many of the customer experience and efficiency challenges facing the financial services industry (FSI). Although the technology has only recently emerged, it is expected to evolve rapidly with improved algorithms, enhanced explainability, wider data access, computing power enhancements and much more. Examples of the pivotal role GenAI will play in FSI include:

• Products and services innovations: GenAI will lead to the creation of new products and services not previously envisioned. For example, it can be used to create personalized investment portfolios, generate personalized insurance policies and create new mortgage options.
• Enhance customer experience: GenAI has the potential to significantly improve the customer experience. For example, it can be used to create smarter chatbots to address complex customer questions and design personalized marketing content.
• Push efficiency with automated tasks: GenAI will automate tasks more effectively than traditional AI and other tech. For example, it can be used to process loan applications faster and more accurately, generate and validate regulatory reports and even manage customer accounts.

Of course, these examples represent just a fraction of the possibilities offered by GenAI, as the industry is still figuring out how to leverage the technology to reach its full potential.

Barriers challenging GenAI adoption in FSI

While GenAI offers exciting possibilities, implementing it across the sector will require organizations to overcome many barriers. A few of the most prominent challenges include:

• Data privacy and security: GenAI runs on large amounts of data, and there is a lack of consistency across GenAI-based AI solutions with respect to how data privacy and security are handled. Financial institutions are mindful of the implications and need to implement strong measures to protect customer and organizational data.
• Regulation and compliance: Regulatory agencies continue to try to catch up with new technologies, including GenAI, the use of which is likely to be subject to regulation. FSI is already a heavily regulated industry, and organizations will need to comply with all applicable GenAI-related regulations — even as they continue to evolve.
• Unspoken costs: With all the excitement surrounding GenAI and its potential to transform the industry, it’s easy to lose sight of the fact that it is expensive to develop and implement. Institutions must consider the costs and benefits to make the right business decisions while also avoiding being left behind as the competition inevitably marches forward.
• Complexity, such as bias: There is always a risk that GenAI models have biases baked into the data used in training and testing the model. Financial institutions will need to address this and other similar complexities (e.g., data availability, hallucination risks, limited explainability, etc.) to adopt GenAI effectively.

Additional barriers include talent availability, cultural nuances and organizational dynamics, to name a few. However, businesses across FSI must focus on eliminating — or at least minimizing — the industry’s longstanding resistance to sweeping technology changes. Its gradual approach to cloud adoption, for example, has left organizations throughout the sector scrambling to catch up to their more technology-progressive competitors. The most successful businesses won’t make that same mistake with GenAI.

And speaking of cloud, let’s explore why and how cloud is an effective answer for accelerating GenAI implementations.

Effective cloud strategy and adoption accelerates GenAI implementation

If implemented effectively, cloud can help financial services organizations drive their digital transformation priorities, improving customer experience, offering competitive products and services and being more efficient. GenAI implementations are far more effective with a solid cloud foundation, thus giving an edge to businesses with higher cloud adoption maturity levels. In fact, industry analysts report that organizations with high cloud maturity are already focusing on bringing GenAI into production, while others are just beginning to define their GenAI roadmaps.

Here are four key ways a strong cloud foundation will help organizations adopt GenAI faster and better:

• Scalability and reliability: GenAI models are computationally expensive to train and run. Strong cloud maturity delivers scalable infrastructure for these models while ensuring that issues like cost controls, performance, security, ease of use, etc., are not barriers to entry (e.g., Google Cloud offers cost-competitive high-performance computing solutions).
• Purpose-ready data strategy: GenAI models need large amounts of data to train. An effective cloud-based data strategy makes it easier to store, process and leverage enterprise data with strict data privacy and applicable regulations for the data. Cloud platforms already provide advanced data preprocessing capabilities and have been further improved to address challenges like bias and performance in their models (e.g., Google Cloud offers vital model evaluation tools to help identify performance and other issues).
• ML ops and AI tooling: Cloud providers have explicitly focused on boosting their platforms to support AI implementation. Effective implementations have led organizations to trust their ML ops and move faster with AI innovations. Challenges such as AI outcome explainability are being addressed through targeted data preprocessing, effective algorithm selections with higher interpretability, features attributions, etc. (e.g., Google Cloud offers Vertex AI platform with capabilities such as model repository, training, evaluation, deployment, monitoring and others).
• Talent dependencies: GenAI is a complex and rapidly evolving space. Specialized talent is not easily available. Cloud implementations are offering ways to address this challenge by simplifying the usage of GenAI. Cloud platforms offer pre-trained models to get started. Data marketplaces are being expanded to offer data to train models. There are no-code/low-code solutions being offered for building GenAI applications (e.g., Google Cloud has recently introduced a model garden for pre-trained models and a GenAI studio for no-code/low-code ways of building GenAI apps).

Additional ways cloud enables GenAI adoption include enabling collaboration and sharing and innovation agility. And while FSI moved slowly on cloud adoption for concerns around security, compliance, cost management and others, GenAI offers a strong reason for businesses across the sector to revisit their cloud strategies and implementation roadmaps to be better prepared to adopt GenAI and other emerging AI technology.

Realizing GenAI value with cloud maturity progression is critical

For the nature of business itself, Financial services, despite the slow adoption of the Cloud, continues to be the industry that may see the most significant value from GenAI adoption. This also highlights the need for the industry to break the resistance to cloud adoption and accelerate the implementation of cloud across their legacy and emerging apps portfolios. Below are several recommendations regarding the priorities that FSI organizations need to focus on to realize GenAI value while they also achieve improved cloud maturity:

• Cloudify data strategy: Bring in the ability to manage large data better and be ready to scale up or down data workloads. Expand analytics capabilities on enterprise data to get better value from it. Enhance collaboration capabilities around data and models, including leveraging data marketplaces where relevant. Strengthen data security with strong encryption methods available. GenAI needs a solid data platform to deliver desired results. Organizations can decide on data deemed fit for cloud; however, they should look at maximizing data on Cloud to get increased value from data. Consider the right cloud platform, design data workloads for cloud, consider cloud-native capabilities, start small and expand.
• Modernize user journey Apps: Enhancing customer experiences and boosting product innovations with GenAI requires a deep look at the user journeys. It’s common for financial services organizations to hit challenges in transforming user journeys for legacy tech stack and tech debts. Consider modernizing user journey-related apps/workloads with cloud on priority. Target embedding GenAI with small pilots initially and expand with increased modernization initiatives over cloud.
• Strengthen AI tooling: To be able to prioritize business transformation, focus on strengthening ML ops and AI tooling with cloud. Cloud platforms can accelerate AI innovations with abilities like pre-trained models, model training accelerators, monitoring capabilities and others. These capabilities are rapidly evolving across cloud platforms. GenAI implementation needs organizations to have the base AI and ML ops framework performing effectively.

Financial services organizations are experiencing a hyper-competitive environment already. Emerging fintechs are being stood up over cloud only to have the competitive edge where they can demonstrate agility and innovation at a breakneck pace. The benefits of cloud for GenAI outweigh the risks and prepare organizations to more effectively and smoothly ride the next innovation wave. To reiterate, the need for FSI to accelerate cloud adoption has never been greater, as the emergence of GenAI creates new opportunities for the industry.