The year 2023 witnessed a surge in discussions around artificial intelligence, fueled by OpenAI’s groundbreaking release of ChatGPT in November 2022. The ensuing excitement reverberated through the stock market, with tech giants like Apple, Microsoft, Google, Amazon, Meta, Tesla, and Nvidia, collectively known as the Magnificent Seven, leading the charge. Nvidia, in particular, dazzled investors with an astonishing 239% return in 2023, exemplifying the burgeoning influence of AI in shaping market dynamics. At the heart of this transformative wave lies generative AI, a special subset poised to redefine the technological landscape.
Generative AI, encapsulated by platforms such as ChatGPT, possesses the unique ability to create diverse forms of data, including text, images, audio, and video. A non-generative model may be well-suited for a task such as facial recognition where, in contrast, generative AI could be used to produce an image of a face solely based on a textual description. It operates by discerning patterns within a dataset and leveraging this knowledge to generate entirely new, yet similar, data with comparable attributes.
The ascent of generative AI gained momentum in 2014, marked by the introduction of Generative Adversarial Networks (GANs) by Ian Goodfellow. GANs, incorporating neural networks, play a pivotal role in tasks such as image generation. More recently, the advent of transformers, as exemplified by the “GPT” in ChatGPT (Generative Pre-trained Transformer), has redefined the landscape of generative AI models. Transformers utilize attention mechanisms, a mathematical approach to identifying data connections, enabling them to filter out irrelevant information and enhance overall efficiency.
The implications of transformer models are profound, ushering in a semi-supervised learning paradigm. This eliminates the need for extensive human labeling of training data, making tasks like image classification more scalable and accessible. Large-language models (LLMs), a subset of generative AI utilizing the transformer architecture, have emerged, featuring prominently in tools like ChatGPT, Microsoft’s Bing Chat, and Google’s Bard. These platforms enable users to input text prompts and receive contextually relevant responses, showcasing the vast potential of generative AI in diverse applications such as content generation, computer programming assistance, summarization of broader texts or topics, language translation, etc. Users can even mimic the style of prominent authors. To showcase the power of this technology, we asked ChatGPT to write a research report on Apple stock in the style of William Shakespeare. Below is an excerpt from the beginning of this masterpiece:
A Bard’s Ode to the Apple Inc. (AAPL) Stock: A Theatrical Rendition
To buy or not to buy, that is the question –
Whether ‘tis nobler in the portfolio to suffer
The slings and arrows of outrageous market volatility,
Or to take arms against a sea of financial troubles
And, by opposing, end them.
This re-characterization of Hamlet’s eminent soliloquy goes on for eight more stanzas, although it may require an English teacher to discern whether it contains a buy or sell recommendation for Apple.
An image of William Shakespeare on Wall Street produced using generative AI within the graphic design tool Canva. A closer inspection of the fine details displays the current shortcomings of generative AI technology.
Of course, it is essential to approach generative AI with caution, acknowledging that while LLMs can provide relevant responses, accuracy is not guaranteed. The potential for misinformation requires users to diligently fact-check outputs, especially when dealing with critical domains like legal research. A law firm made headlines last year when it used ChatGPT for assistance in putting together a legal brief which was filed with the court. ChatGPT was able to identify several court cases with relevant precedent which were included in the brief. Yet, there was a complication: some of the cases being cited as precedent did not exist; they were created by ChatGPT.
In the investment industry, generative AI may assist in tasks such as research summaries, sentiment analysis, identifying key points of management calls, and summarizing analyst opinions. On a more technical level, quantitative researchers can employ generative models to supplement observed data, thereby increasing the volume of data used in training specialized machine learning models. Ultimately, generative AI is not a substitute for robust investment management. Generative AI’s true value lies in augmenting human decision-making rather than supplanting it.
In conclusion, generative AI represents a pivotal force in technological advancement, offering a tantalizing glimpse into a future where human-machine collaboration reaches unprecedented heights. From content creation to problem-solving, the capabilities of generative AI pave the way for innovation across diverse industries. As we navigate this exciting frontier, it is imperative to strike a balance between embracing the potential and responsibly harnessing the power of generative AI.
Author’s Note: For those wondering, yes, ChatGPT was used in the production of this article. The first draft was written by the inferior human hand, and this was given to ChatGPT with the prompt: “Take the following blog article and improve the writing style.” Some additional changes were made for the final version.
The views expressed represent the opinion of Passage Global Capital Management, LLC. The views are subject to change and are not intended as a forecast or guarantee of future results. This material is for informational purposes only. It does not constitute investment advice and is not intended as an endorsement of any specific investment. Stated information is derived from proprietary and nonproprietary sources that have not been independently verified for accuracy or completeness. While Passage Global Capital Management, LLC believes the information to be accurate and reliable, we do not claim or have responsibility for its completeness, accuracy, or reliability. Statements of future expectations, estimates, projections, and other forward-looking statements are based on available information and Passage Global Capital Management, LLC’s views as of the time of these statements. Accordingly, such statements are inherently speculative as they are based on assumptions that may involve known and unknown risks and uncertainties. Actual results, performance or events may differ materially from those expressed or implied in such statements.