Intelligent Document Processing and its Potential in Financial Services

Marco Argenti, CIO at Goldman Sachs, discusses the use cases of generative AI in financial services, specifically focusing on intelligent document processing. Financial institutions deal with numerous documents, such as contracts, filings, and communications, and transforming them into machine-readable structures is a challenge. Machine learning techniques, like entity extraction and natural language processing, are used to convert documents into structured data. However, the limitations of generative AI, such as potential errors, require techniques to ensure information accuracy in document processing. By leveraging vector databases and retrieval augmented generation, along with training users to write effective prompts, the future of document management can combine human and machine capabilities for increased productivity and reduced errors.