AI’s Limits Exposed: Why Human Judgment Still Reigns in Stock Market and Baseball
Artificial intelligence (AI) continues to make waves in various industries, including the stock market and baseball. However, it is essential to understand the limitations of AI and recognize the significant role human judgment plays in decision-making.
In recent times, AI has garnered attention and influenced stock market behavior. Senior executives at Standard & Poor’s 500 companies mentioned AI or artificial intelligence an average of 3.7 times per call with analysts in the second quarter of 2023, compared to 1.8 mentions per call in the previous quarter. Additionally, stocks focused on AI, such as Upstart Holdings Inc., experienced significant surges, gaining over 400% in just a few months before experiencing a decline.
The rise of AI has also raised concerns among employees about job security. During the Writers Guild of America strike, AI threatening job positions was a major issue. While AI brings both excitement and dread, it is crucial to acknowledge its current limitations. AI excels in making predictions but falls short when it comes to making judgment calls that involve personal preferences and experiences.
A recent article in the Harvard Business Review aptly summarizes this point, stating that AI is a prediction machine. It can inform us about the probability of rain but cannot determine whether we should pack an umbrella. The umbrella decision requires more than just prediction; it necessitates judgment based on individual preferences.
This interplay between prediction and judgment is also evident in the world of investments. Different individuals have varying risk tolerances, which dictate their investment choices. While the probability of gain or loss from certain investments may be the same for all investors, their preferences and tolerances for risk differ. Determining an investor’s risk tolerance is not a straightforward task.
A similar dynamic can be observed in the world of baseball. The playoffs often showcase the delicate balance between relying on AI-backed data and trusting human judgment. Some teams tend to replace starting pitchers early in games based on data indicating reduced effectiveness when facing batters for the third time. While this decision may seem logical, it deprives starting pitchers of the opportunity to build mental confidence and develop a repertoire of deceptive pitches – qualities typically associated with successful pitchers in the past. Furthermore, it can also strain relief pitching staff, potentially costing games in the long run.
The playoff experiences of Jordan Montgomery, the star pitcher for the Texas Rangers, highlight these concerns. Montgomery won three starts in the playoffs, despite the New York Yankees trading him the previous year due to doubts about his ability to excel in big postseason games. The decision to trade him was influenced by data-driven insights that favored early pitcher substitutions. However, this decision disregarded the potential for individual pitchers to navigate challenging situations and improve their performance as the game progressed, as exemplified by legendary pitchers like Tom Seaver and Fernando Valenzuela.
These examples demonstrate the limitations of prediction machines. AI lacks the ability to factor in individual human preferences, experiences, and the capability to learn, adapt, and adjust on the fly. In the business world, these limitations can have costly implications, as evidenced by the real estate company Zillow Group Inc. Its AI-powered solution to value homes initially seemed promising, but the company suffered a substantial loss when it had to take a $304 million inventory writedown due to overpaying for properties. This event resulted in plummeting stock prices and staff reductions.
Despite the undeniable impact of AI, it should not replace the need for human judgment. While AI can provide valuable insights based on data, it may lack the empathy required in decision-making and delivery. Augmented intelligence – the combination of AI and human judgment – offers a more balanced and effective approach.
In conclusion, AI has undoubtedly revolutionized various industries, but it has its limits. It cannot replicate the nuances of human judgment, individual preferences, and the ability to adapt in real-time. Therefore, a careful balance between AI and human judgment is necessary to make informed decisions that consider both data-driven insights and human experiences. As the entrepreneur Gary Vaynerchuk emphasizes, If content is king, then context is God. AI has the potential to master content, but mastering context remains a work in progress.