How AI is transforming investing

Jun 15, 2023

Key points

  • The release of generative artificial Intelligence (“AI”) platforms like ChatGPT has driven increased attention around AI and the latest developments in natural language technology.
  • Similar to how these AI platforms use transformer technology to create large human-like text outputs, this technology can be used to help inform investment decisions.
  • Within BlackRock Systematic, we use transformers to maximize the accuracy and precision of natural language processing (“NLP”) across a wide range of data sources—uncovering potentially valuable investment insights.

Artificial Intelligence (“AI”), or the simulation of human intelligence by machines, has been evolving for decades. Generative AI is the latest breakthrough category in the space, garnering attention for its ability to create original ideas and content. Near the end of 2022, OpenAI released its AI platform, “ChatGPT” to the public. ChatGPT uses advanced language technology to create large human-like text outputs—bringing the most recent advancements in generative AI to the masses. ChatGPT exploded in popularity at the fastest pace of any online application, reaching one million users in just five days (Figure 1).

Figure 1: ChatGPT reached one million users faster than any other online application
Amount of time to reach one million users for online applications

Chart shows the length of time that online applications have taken to reach one million users. ChatGPT, OpenAI’s artificial intelligence tool, only took five days to reach one million users, making it the fastest of any online application.

Source: Statista, with data from company announcements via Business Insider/LinkedIn, as of January 24, 2023. Kickstarter measured as one million backers, Airbnb measured as one million nights booked, Instagram measured as one million downloads.

So what is the technology underpinning this disruptive platform? ChatGPT is a large language model (“LLM”) based on generative pre-trained transformer (“GPT”) technology. LLMs are trained using massive amounts of data sourced from websites, books, academic publications, and other public datasets. These models are trained to predict the next word in a text given previous context, and in that process they acquire linguistic skills, world knowledge, as well as basic reasoning skills. Coincidentally, on the same day that Chat GPT was released to the public, BlackRock Systematic approved an investment insight that leverages the same transformer technology powering LLMs. Let’s explore how we use these models to enhance our investment capabilities.

Investing in the era of AI

For decades, we’ve been applying natural language processing (“NLP”) techniques across a wide range of text sources including broker analyst reports, corporate earnings calls, regulatory filings, and online news articles. When analyzed at scale, each individual insight can be combined into an aggregate view that helps inform our return forecasts. The more effectively we’re able to extract and understand these insights, the more of an investment edge they may be able to provide.

Early investment signals used sentiment analysis, tracking the number of positive and negative words included in a document, and assigning an overall sentiment score based on word counts. While these signals proved effective, they weren’t designed to factor in nuances like sentence structure and semantics that can impact the meaning of text. New research innovations have improved the granularity of text analysis over time. For example, advancements in machine learning helped to determine the most relevant words to track based on the type of text input.

Today, our approach has evolved to utilize transformer-based LLMs (just like ChatGPT). Transformers are a type of neural-network architecture that can process long sequences of elements (like words in a sentence), accounting for the relationship between each individual word with other words and focusing on the most important points (Figure 2). This differs from other methods of text analysis that are limited to processing information sequentially and tend to overemphasize neighboring words—potentially missing important connections between words that are more distantly separated. Because of this, transformer-based models tend to provide a more accurate and precise understanding of the text. Just like ChatGPT can use this technology to predict the next word in a sentence and produce human-like content, we can leverage it to improve our investment predictions.

Figure 2: The transformer architecture allows for a more accurate analysis of text by considering the interactions between words in a sentence and identifying the most significant relationships
Hypothetical example of text analysis using transformer architecture

Image illustrates how transformer architecture enables more accurate text analysis by measuring interactions between words and identifying the most important relationships. This image shows the varying levels of relevance between the word “company” and other words in the sentence.

Source: BlackRock Systematic, for illustrative purposes only, as of June 2023.

LLMs are trained using a vast number of data inputs. This is what allows ChatGPT to perform a wide range of tasks and closely simulate human reasoning with broad applicability. By comparison, the LLMs used in our investment process are designed to complete specific investment tasks, such as forecasting the market reaction following corporate earnings calls, for example. As a result, our models are trained on a smaller set of data inputs but are expected to deliver a high level of accuracy in performing the specific task that they’ve been trained and fine-tuned for. Figure 3 illustrates the performance of our earnings call model compared to OpenAI’s larger GPT models at predicting post-earnings market reactions. While the accuracy of OpenAI’s model improved considerably from GPT-3.5 to GPT-4, both OpenAI models demonstrate a lower level of predictive performance than our proprietary model that’s been trained and fine-tuned for this particular purpose.

Figure 3: The BlackRock Systematic earnings call model has been fine-tuned to predict post-earnings market reactions with a high level of accuracy
Accuracy of models at predicting post-earnings market reactions

Image shows the level of accuracy of a Systematic Model designed to predict the market reaction following quarterly earnings calls vs. OpenAI’s GPT model. This illustrates how when used in investing, our models are trained on a smaller set of data inputs but are expected to deliver a high level of a

BlackRock Systematic, as of May 2023. This analysis is based on a sample of 200 earnings calls. The analysis computes the prediction for each model and compares it with return outcomes (positive or negative) based on future 3-day stock returns. The accuracy is computed as the fraction of predictions that were correct for each model.

Identifying AI winners

Along with transforming the way we invest, AI is impacting the investment opportunity set. Figure 4 shows the performance of a proprietary investment insight that’s designed to capture the winners in the new era of AI. So far, we’re seeing the first-order effects of AI being priced in as markets reward a small subset of AI innovators while punishing their more traditional media counterparts. Over time, we expect ChatGPT and related technologies to act as a catalyst for more widespread integration of AI in business models across industries.

Figure 4: A small subset of AI innovators have driven outsized gains in the tech space
Cumulative returns of AI investment insight

This chart shows the performance of an investment insight designed to identify the companies positioned to benefit from new AI technologies. Currently, the market is rewarding a small subset of AI innovators.

BlackRock Systematic, as of May 2023. The chart shows companies within the technology sector defined by BlackRock Systematic as having exposure to Artificial Intelligence technologies. This is for illustrative purposes only and is not meant to represent the past or future performance for the sector shown.

The cutting-edge of innovation in investing

As systematic investors, we focus on generating alpha by maintaining an information advantage in markets. We’ve been researching and integrating AI, machine learning, and NLP technologies into our investment process for many years.1 The recent release of ChatGPT brought the latest breakthroughs in AI to the public’s awareness, driving an explosion of excitement and attention around the topic. What may be less well known is that the predictive abilities of AI can also be applied in the investing world. Within BlackRock Systematic, these technologies enhance our ability to analyze datasets and forecast investment outcomes—transforming the way we invest by remaining on the cutting-edge of innovation.

Raffaele Savi
Raffaele Savi
BlackRock Systematic Active Equity Investment Team
Jeff Shen, PhD
Jeff Shen, PhD
Co-Head of Systematic Active Equity
Yaki Tsaig
Research Scientist, Systematic Active Equity