How can financial services firms derive value from generative AI?

In the ever-evolving landscape of artificial intelligence, large language models (LLMs) have captured the world’s attention and ignited a revolution in language understanding and generation. For example, ChatGPT gained more than 100 million monthly active users in less than three months, making it the fastest growing application in history. These remarkable advancements stand at the forefront of generative AI, pushing the boundaries of what machines can do with text and language.

The adoption of LLMs has sparked discussions and varying viewpoints among industry observers, academia, regulatory bodies and the public at large. Governments are being urged to accelerate AI regulation in response to the widespread use of generative AI models and the associated risks they present. Major concerns center around privacy, trust and security.

Despite these concerns, generative AI is widely believed to be a lasting technology that will transform ways of working for numerous industries. The level of expenditure in AI by corporations has been rapidly increasing, as all industries are investing significant time, money and resources in actively evaluating this technology. To that end, some are focused on more controlled experimentation, while others have announced a multiyear commitment of embedding this technology across enterprise use cases.

Value proposition for financial services

While banks and financial institutions have been deploying artificial intelligence (AI) applications across a variety of use cases for several years, including managing credit risk and fraud, generative AI in financial services represents a step change from previous approaches. It holds the potential to revolutionize a much broader array of business functions.

Numerous applications have been identified as ripe for potential use, among them redefining the future of financial advice, insurance claims processing, customer marketing, engagement and servicing. Internal applications such as compliance monitoring, contact center operations, application development and maintenance are also in consideration. The expansion of use cases is driven in part by the significant advancements in the capabilities offered by this technology, particularly in parsing and making sense of unstructured data such as text. Even more potential uses are enabled by the ability to query data in a natural human interaction or Q&A format and provide natural language instructions to create or refine new business content.

At this time, however, the technology is still in its early stages, and generative AI must be used in an ecosystem where its combination with human expertise fosters synergy that leverages the strengths of both to deliver accurate insights and generate value. The primary value areas, acknowledging the current strengths and drawbacks of this technology, fall under two primary categories:

• Operational efficiency: Enhancing productivity and reducing costs by automating routine tasks such as information review, comparison, categorization and synthesis.
• Augmented intelligence: Assisting human experts by providing insights, recommendations, creative content and decision-making support.

These areas also enable various second-order effects such as better client experience through timely and adequate support, focusing human effort on more intellectually challenging tasks while streamlining other activities.