Generate an Equity Research Report on Any Stock
Beta test version
Last updated: 20 October 2025
Objective:
Sometimes we want to push AI to make a call on a stock to see whether its synthesis of data leads it to a similar conclusion to our own - or to play devil’s advocate and challenge our thinking. This prompt leads the AI though an assessment of recent results, guidance, growth, news, valuation and bull and bear investment theses, to synthesize what it believes (probabilistically) is the likely outlook for a stock over the next 12 months.
Disclaimer: Use this prompt cautiously to test your knowledge and logic on a stock but not as a predictor of the future. This is not intended to be financial advice and no recommendation, either express or implied, is made on any stock. The AI’s accuracy on stock market predictions is untested, it can and does make mistakes and utilize incorrect data. Any forward looking view on a stock contained in AI outputs, do not reflect the opinions of the Inferential Investor. Seek professional financial advice before making any investment decisions.
Explanation:
This prompt is a beta test to see how close to Wall St broker research we can get using existing AI models and retrieval augmented generation (web scaping of data). Over time we expect these tasks to get better and better as model capabilities develop.
We build into the context window, layers of information: recent results, growth, guidance, earnings revisions, valuation etc from which the model is then asked to synthesize an investment thesis and outlook.
The model also retrieves a synopsis of both the bull and bear case narratives for the stock to compare against its thesis as well as publicly available broker price targets. It compares its conclusions against these scenarios, highlighting divergent assumptions and measures its own confidence in the conclusion. Reliant as it is on web based RAG, expect some data to be missed or incorrectly retrieved and this should be corrected in a chain of thought conversation with the model updating its conclusions following this input.
The larger and better covered the stock (news and research), the better the quality of outputs from this approach as the prompt is not linked to specific financial database APIs. It is suggested that a 2 model approach is used to draw out more information and sense check conclusions.
As always, be aware that models can make mistakes. At each step, examine the response and challenge information or conclusions that appear erroneous before proceeding to any subsequent steps. If in doubt use a second model with the same prompt to verify the information and generate challenge questions and answers (CoVe process) to correct interpretations of data.
Link to blog post explanation:
N/A
Preferred Model(s):
Dual model technique recommended. Gemini 2.5+ (DEEP RESEARCH MODE) and ChatGPT-5+ with Chain-of-Verification (CoVe) follow up in each.
Important Execution Notes:
Insert the stock ticker, exchange and name where indicated.
Utilize multiple models to draw out more information and chain of thought prompting to add considerations to each that the other model has surfaced.
It is recommended to also use CoVe procedures to verify data further. Request the model to generate a list of verification questions, then get the same model to answer those and update the report accordingly to see if conclusions change.
Sample Output:
Copy/Paste Prompt Set:
Important note: Subscribers can use this prompt set for their own analysis. However, the prompt is copyrighted by The Inferential Investor, paywalled, and must not be shared without permission.