The following essay is reproduced with permission from 
The conversationan online publication covering the latest research.
AI-based tools, such as ChatGPT, have the potential to revolutionize the efficiency, effectiveness, and speed of work done by humans.
And that goes as much for the financial markets as it does for sectors like health care, manufacturing and just about every other aspect of our lives.
I researched the financial markets and algorithmic trading for 14 years. Although AI offers many advantages, the increasing use of these technologies in financial markets also points to potential perils. A look at past Wall Street efforts to accelerate trading by embracing computers and AI offers important lessons about the implications of their use for decision-making.
Program trading fuels Black Monday
In the early 1980s, fueled by technological advancements and financial innovations such as derivatives, institutional investors began to use computer programs to execute trades based on predefined rules and algorithms. This helped them to complete large transactions quickly and efficiently.
At the time, these algorithms were relatively simple and were mainly used for so-called index arbitragewhich consists of trying to take advantage of the differences between the price of a stock market index – such as the S&P 500 – and that of the stocks that make it up.
As technology advanced and more and more data became available, this type of trading program became increasingly sophisticated, with algorithms capable of analyzing complex market data and executing transactions based on a wide range of factors. These program traders continued to grow in number on the largely unregulated trade highways – on which more than one billions of dollars in assets change hands every day – causing market volatility will increase significantly.
Eventually, this led to the massive stock market crash in 1987 known as Black Monday. The Dow Jones Industrial Average suffered what was at the time the largest percentage decline in its history, and the pain spread around the world.
In response, regulators put in place a number of measures to restrict the use of program exchanges, including circuit breakers that interrupt exchanges in the event of significant market fluctuations and other limits. But despite these measures, program trading continued to grow in popularity in the years following the crash.
HFT: trading programs on steroids
Fast forward 15 years to 2002 when the New York Stock Exchange introduced a fully automated trading system. As a result, program traders gave way to more sophisticated automations with much more advanced technology: High frequency trading.
HFT uses computer programs to analyze market data and execute trades at extremely high speeds. Unlike program traders who have bought and sold baskets of securities over time to take advantage of an arbitrage opportunity – a difference in price of similar securities that can be exploited for profit – high frequency traders use powerful computers and high-speed networks to analyze market data and execute trades at lightning-fast speeds. High Frequency Traders can complete transactions in approximately one 64 millionth of a secondcompared to the seconds it took traders in the 1980s.
These transactions are usually very short term and may involve buying and selling the same security multiple times within nanoseconds. AI algorithms analyze large amounts of data in real time and identify patterns and trends that are not immediately apparent to human traders. It helps traders make better decisions and execute trades at a faster rate than would be possible manually.
Another important application of AI in HFT is natural language processing, which involves the analysis and interpretation of human language data such as news articles and social media posts. By analyzing this data, traders can gain valuable insight into market sentiment and adjust their trading strategies accordingly.
Benefits of AI Trading
These AI-powered high-frequency traders work very differently from people.
The human brain is slow, imprecise and forgetful. It is incapable of performing fast, high-precision floating-point arithmetic needed to analyze huge volumes of data to identify trading signals. Computers are millions of times faster, with essentially foolproof memory, perfect attention, and unlimited capacity to analyze large volumes of data in milliseconds.
Thus, like most technologies, HFT offers several advantages to stock markets.
These traders usually buy and sell assets at prices very close to the market price, which means that they do not charge high fees to investors. This ensures that there are always buyers and sellers in the market, which helps to stabilize prices and reduce the risk of sudden price fluctuations.
High-frequency trading can also help reduce the impact of market inefficiencies by quickly identifying and exploiting market pricing errors. For example, HFT algorithms can detect when a particular stock is undervalued or overvalued and execute trades to take advantage of those deviations. By doing so, this type of trading can help correct market inefficiencies and ensure that assets are priced more accurately.
The inconvenients
But speed and efficiency can also cause damage.
HFT algorithms can react so quickly to news events and other market signals that they can cause sudden spikes or drops in asset prices.
Additionally, HFT financial companies can use their speed and technology to gain an unfair advantage over other traders, further distortion of market signals. The volatility created by these extremely sophisticated AI-powered trading beasts led to the so-called flash crash in May 2010, when stocks plunged then recovered in minutes – erasing then restoring approximately $1 trillion in market value.
Since then, volatile markets have become the new normal. In a 2016 research, two co-authors and I found that volatility – a measure of how quickly and unpredictably prices go up and down – increased significantly after introduction of HFT.
The speed and efficiency with which high-frequency traders analyze data means that even a small change in market conditions can trigger a large number of trades, leading to sudden price swings and increased volatility.
Besides, research I have published with several other colleagues in 2021 shows that most high-frequency traders use similar algorithms, which increases the risk of market failure. Indeed, as the number of such traders increases in the market, the similarity of these algorithms can lead to similar trading decisions.
This means that all high frequency traders could be trading on the same side of the market if their algorithms emit similar trading signals. In other words, they could all try to sell on negative news or buy on positive news. If there is no one to take the other side of the trade, markets can fail.
Enter ChatGPT
This brings us to a whole new world of ChatGPT powered trading algorithms and similar programs. They could take the problem of too many traders on the same side of a deal and make it even worse.
In general, humans, left to their own devices, will tend to make a wide range of decisions. But if everyone derives their decisions from a similar artificial intelligence, it can limit the diversity of opinions.
Consider an extreme, non-financial situation where everyone depends on ChatGPT to decide on the best computer to buy. Consumers are already keen to herd behavior, in which they tend to buy the same products and models. For example, reviews on Yelp, Amazon, etc. encourage consumers to choose from some of the best choices.
From the decisions made by the generative AI-powered chatbot are based on past training data, there would be a similarity in the decisions suggested by the chatbot. Chances are that ChatGPT would offer the same make and model to everyone. This could take livestock farming to a whole new level and lead to shortages of certain products and services as well as sharp price spikes.
This becomes more problematic when the AI making the decisions is informed by biased and incorrect information. AI algorithms can reinforce existing prejudices when systems are trained on biased, old or limited datasets. And ChatGPT and similar tools have been criticized for making factual errors.
Also, since stock market crashes are relatively rare, there is not much data about them. Since generative AIs depend on data training to learn, their lack of knowledge about them could make them more likely to occur.
For now, at least, it looks like most banks won’t allow their employees to take advantage of ChatGPT and similar tools. Citigroup, Bank of America, Goldman Sachs and several other lenders have already banned their use in trading rooms, citing privacy concerns.
But I strongly believe that banks will eventually adopt generative AI, once they solve the problems they have with it. The potential gains are too great to ignore – and there is a risk of being left behind by rivals.
But the risks to financial markets, the global economy and everyone else are also great, so hopefully they will tread carefully.
This article was originally published on The conversation. Read it original article.