Training GPT3 to be an Investment Banker

Hey PH community, We have been experimenting with using GPT3's reasoning capabilities to enhance our finclout.io platform to simulate the work of a junior equity analyst in an investment bank. While it might be a bit early for Goldman Sachs to start shaking in fear, the results were quite impressive. We started by prompting GPT3 to perform a few simple tasks: (1) Summarize relevant news of the last 30 days, (2) Write a SWOT analysis, (3) Identify competitors, (4) Identify Key Risks, and (5) Write three investment theses (bull, neutral, bear) The workflow is quite straightforward. All connectivity with GPT3 was handled through the OpenAI API. Then we combined the returned datasets with our existing hyperfocus data and presented them to a selected group of test users on finclout.io in the form of visually pleasing dashboards. (We currently have about 7K unique monthly users) Here are some examples of how these dashboards look like: https://app.finclout.io/tp/MSFT https://app.finclout.io/tp/LYFT https://app.finclout.io/tp/AAPL https://app.finclout.io/tp/UBER https://app.finclout.io/tp/BBBY Key takeaways: The prompts need to be well-engineered to return a reliable result. Pro-tip: Embed the GPT3 API calls in prompts that return json as datatypes in a pre-defined format. The returned data especially around news summarization, competitor analysis, and key risk was really impressive. One notable downside of this integration was surprisingly API rate limits. Even though we pay for the service, have waiting periods between calls, and only allow certain users to trigger the call we frequently didn't get data back because of rate limits. In some cases, we observed obvious errors. "Uber Technologies, Inc. reported a surge in fourth-quarter revenues and a strong outlook for 2021." We are not in 2021 anymore for a long while. The reason for this error could be that GPT3's knowledge of the Yahoo finance dataset has been cut off in 2021. In conclusion, while we need to further investigate if the actual reasoning of GPT3 is mature enough to provide a useful analysis of the stock the results are already quite promising. One potential pitfall will not be a legal problem but rather a regulatory one. What happens if GPT3 returned factually incorrect information leading to a liability exposure. Clearly, that is only an experiment and shouldn't be used as an investment recommendation nor seen as investment advice in any way. But if the results are consistent and reliable, it could be a real game changer.