The study suggests that LLMs could play a central role in decision-making due to their vast knowledge base and ability to recognize patterns and business concepts. Despite challenges in numerical analysis, the ability of a general-purpose language model like GPT-4 to match specialized ML models and exceed human experts indicates the potential disruptive impact of LLMs in the financial domain.
Key takeaways:
- Researchers from the University of Chicago have found that large language models (LLMs), specifically OpenAI's GPT-4, can conduct financial statement analysis with accuracy rivaling and even surpassing that of professional analysts.
- The researchers used a novel approach of providing structured financial data and “chain-of-thought” prompts to guide the AI’s reasoning, achieving a 60% accuracy in predicting the direction of future earnings.
- The study suggests that LLMs may take a central role in decision-making due to their vast knowledge base and ability to recognize patterns and business concepts.
- Despite challenges in numerical analysis, the ability of a general-purpose language model to match the performance of specialized ML models and exceed human experts points to the disruptive potential of LLMs in the financial domain.