However, skeptics question the validity of these claims. Critics argue that the artificial neural network model used for comparison is outdated and cannot be compared to current advanced models. AI researcher Matt Holden also questioned the model's ability to pick stocks that can outperform a broader index like the S&P 500. Despite skepticism, the researchers believe that large language models can be powerful tools to aid human analysts in their work.
Key takeaways:
- OpenAI’s GPT-4 large language model has demonstrated an ability to analyze financial statements with a level of accuracy that surpasses the best human financial analysts, according to a paper written by researchers at the University of Chicago.
- The researchers used a technique known as “chain-of-thought” prompting to enable GPT-4 to undertake more complex reasoning, and found it could predict the direction of future earnings with 60% accuracy, surpassing the 53% to 57% accuracy of most human financial analysts.
- However, some skeptics question the validity of these claims, pointing out that the artificial neural network model used as a benchmark by the researchers is outdated, and doubting that GPT-4 can outperform a broader index such as the S&P 500.
- Despite skepticism, the researchers believe that large language models can be powerful tools to help streamline the work of human financial analysts, and have created an interactive web application to demonstrate GPT-4’s ability to perform financial analysis.