GPT-4 Outperforms Human Analysts in Predicting Company Profit Growth

According to TapTechNews on May 27, a new study from the University of Chicago shows that the large language model GPT-4 can outperform human analysts in predicting the future profit growth of companies, and this AI model only used the company's financial statements without additional information assistance.

In the past, financial analysts relied on professional knowledge and experience to assess the company's financial situation and predict future profits. However, this study shows that the AI model can complete this task equally well or even better.

As TapTechNews understands, the researchers provided the model with anonymized financial data, including balance sheets and income statements, and required it to predict future profit growth. Even without any additional information, GPT-4 can achieve an accuracy rate of 60%, while the typical accuracy rate range of human analysts is only 53% to 57%.

Part of the reason for GPT-4's success lies in the use of chain of thought prompts, which guide the AI model through the analysis process, helping it identify trends, calculate ratios, and integrate information to make predictions.

The researchers believe that due to its large knowledge base and strong pattern recognition ability, large language models have significant advantages over human analysts.

Although some experts cautiously point out that the benchmark used in this study may not be the most advanced financial analysis model, the overall research results are still encouraging. The general AI model can surpass specialized financial models and even human experts, indicating that large language models have the potential to change the field of financial analysis.