We asked GPT-4 to describe the most investable stock in NYSE and Nasdaq

In this essay, we will describe the technology, methodology, and results.

What is GPT-4?

GPT-4, or Generative Pre-trained Transformer 4, is a state-of-the-art deep learning model developed by OpenAI. It's part of a series of transformer models that utilize large-scale machine learning techniques to generate human-like text based on an input prompt.

The architecture is based on the transformer model introduced by Vaswani et al. in 2017, using self-attention mechanisms to capture contextual information across the input sequence. GPT-4 is trained on a large corpus of text data, and it employs unsupervised learning, allowing it to predict the next word in a sentence given the previous words.

LLMs like GPT-4 are fundamentally predictive engines, optimized to anticipate the most probable subsequent word in a sequence with exceptional accuracy.

Turns out, reasoning can be automated simply by predicting the most probable subsequent word.

Large Language Models like GPT-4 offer the significant advantage of processing information free from emotional influence.

In the current AI landscape, autonomous agents already have a multitude of abilities, including internet browsing, code writing, and pattern recognition. These are invaluable skills, especially when applied to areas such as stock picking.

Prompt/Objective:

We used a two-step approach to generate the stock recommendation.

Step 1: We designed a prompt for GPT-4 (using ChatGPT Plus) to come up with a description of the most investable stock from the NYSE and Nasdaq. Note that GPT-4's training cut-off date is Sept 2021 and currently doesn't support web browsing.

We used ChatGPT Plus for this 1st step.