Exploring the Early Impact and Barriers of Adoption
As generative AI continues to make waves across industries, its potential impact on the global economy is becoming increasingly evident. According to a McKinsey report, generative AI has the potential to add trillions of dollars annually to the global economy, with the banking industry being one of the sectors that could benefit the most. However, amidst the hype surrounding this technology, it is crucial to separate the promises from the reality. In this MIT Technology Review Insights report, we delve into the early impact of generative AI in the financial sector, examining its current applications and the challenges that need to be overcome for successful deployment.
Nascent Deployment in Financial Services
While generative AI is still in its nascent stages of deployment within the financial services industry, some companies have begun to harness its power. The most active use cases currently revolve around automating low-value, repetitive tasks, freeing up employees to focus on more strategic and complex work. By leveraging generative AI tools, financial institutions can automate time-consuming jobs that previously required human intervention, particularly in assessing unstructured information.
Exploring Disruptive Tools
Beyond the automation of mundane tasks, researchers and financial institutions are actively exploring the potential of generative AI in more impactful areas. These include asset selection, improved simulations, and a better understanding of asset correlation and tail risk. However, the commercial deployment of these tools remains rare due to a range of practical and regulatory challenges that need to be addressed.
Legacy Technology and Talent Shortages
Legacy technology and talent shortages pose temporary obstacles to the widespread adoption of generative AI tools in the financial sector. Many financial services companies, especially large banks and insurers, still rely on aging information technology and data structures that may not be suitable for modern applications. However, with the widespread digitalization efforts in recent years, this problem has been gradually alleviated. Another challenge lies in the scarcity of talent with expertise in generative AI. Financial institutions are currently training their existing staff rather than competing for a limited pool of specialists. Nevertheless, the shortage of AI talent is expected to diminish over time, as has been observed with the rise of other new technologies.
Technological Limitations and Regulatory Hurdles
While progress is being made, there are still significant obstacles to overcome in terms of the technology itself and regulatory considerations. Off-the-shelf generative AI tools may not be capable of performing complex, specific tasks such as portfolio analysis and selection. Companies will need to invest time and resources in training their own models to achieve the desired outcomes. Additionally, ensuring the accountability and unbiased nature of generative AI output remains a challenge. Authorities recognize the need for further study of the implications of generative AI and have historically been cautious in approving new tools for widespread use.
Generative AI holds immense potential for the financial sector, with the ability to automate low-value tasks and revolutionize critical areas such as asset selection and risk analysis. While deployment is still in its early stages, the barriers of legacy technology and talent shortages are gradually being overcome. However, challenges related to the technology itself and regulatory considerations remain. As the financial industry continues to explore the possibilities of generative AI, it is crucial to navigate these challenges to ensure successful and responsible deployment.