A recent study conducted by the Booth School of Business at the University of Chicago found that OpenAI's GPT-4 outperformed human financial analysts in analyzing financial statements and making forecasts. The study revealed that even without any narrative or industry-specific information, the large language model (LLM) was better at predicting earnings changes than human analysts.
Utilizing "chain-of-thought" prompts, GPT-4 was able to identify trends in financial statements, calculate financial ratios, and make accurate predictions about future earnings. The study showed that the LLM exhibited a relative advantage over human analysts, particularly in situations where analysts tend to struggle.
The study's authors highlighted the potential for large language models like GPT-4 to play a significant role in financial decision-making due to their ability to recognize financial patterns and business concepts with incomplete information. Additionally, the study found that trading strategies based on GPT-4's forecasts delivered more profitable results than the stock market, with higher share ratios and alpha.
These findings have the potential to disrupt the financial services industry, which is increasingly adopting generative AI technologies. The study's results suggest that GPT-4's financial acumen could be leveraged to improve trading strategies and enhance overall financial decision-making processes.
Overall, the study provides evidence that OpenAI's GPT-4 is a powerful tool for financial analysis and forecasting, outperforming human analysts in certain tasks. As businesses continue to explore the capabilities of AI technologies, findings like these may shape the future of financial services and investment strategies.