When it comes to generative artificial intelligence (AI) and its ability to deliver securities-specific advice to investors, beauty is in the eye of the beholder.
As just one example and for the purposes of this article, I asked ChatGPT this question: Is Costco (NASDAQ: COST) a good stock? That’s a yes/no question and when an investor makes a query of that nature, their hope is the respondent, including of the non-human variety, replies with a straight-forward answer and supporting data and facts.
ChatGPT doesn’t do that with questions of that nature. Its response to my Costco inquiry was impressively detailed, but also heavily hedged. In other words, the AI platform laid out pros and cons regarding Costo stock, but it didn’t provide me with a definitive buy/sell answer.
Likewise, according to my perusals of various Reddit investing forums, if one asks a generative AI tool such as ChatGPT to build investment portfolio, the answers are almost universally something along the lines of four to six plain vanilla ETFs. Don’t expect anything of the tactical variety, let alone something “adventurous” such as a sector or industry fund.
Alright, so ChatGPT and its peers don’t have all the answers some investors are hoping for, but “picking winners” isn’t the point of AI. In fact, generative AI can fill an important void for investors: addressing various forms of cognitive bias.
Beating Bias With AI
It’s human nature to have biases and that’s where AI can be impactful for investors. While AI is taking on more human thinking traits, it’s designed to limit or eliminate bias. It has more self-control than humans and that’s a positive when it comes to making investment decisions. Consider the case of recency bias.
Our minds naturally reach out to what happened most recently when predicting what will happen in the future. In investing, our natural tendency to focus closely on recent events can get us in trouble,” notes Morningstar’s Samantha Lamas. “Thus, it can help to have a ‘thinking partner’ that nudges you to think outside of recent events. ”
Tapping AI to damp recency bias is highly relevant here and now because some of the macroeconomic factors that chased skittish investors to cash and selling low earlier this year could be creeping back into the economic/investing conversation. There’s nothing wrong with considering AI as an avenue for avoiding making the same mistake twice. Plus, AI could be just what the doctor ordered in terms of dealing with temporal discounting.
“Temporal discounting leads us to discount future needs in place of present-day needs and desires. Some research suggests this is because we feel psychologically disconnected from our future selves, to the point that our future selves can be seen as complete strangers,” adds Lamas. “In investing, this can prompt us to ignore our long-term needs in favor of our present-day needs—for example, by making decisions that might make us feel better right now but could jeopardize our financial futures. ”
More Ways AI Can Quash Bias
One of the most prominent (and dangerous) forms of bias its confirmation bias. It’s essentially having an opinion or view the holder doesn’t really want altered even if they’re seeking alternative perspectives. For example, if I had a strong bullish view on Costco prior to investigating the matter further, it may have been hard to convince me otherwise, even if there was compelling evidence to that effect.
That’s an example of how AI can be advantageous when it comes to mitigating biases that can derail an individual investment’s process and results. AI isn’t driving an agenda. It’s not trying to sell us anything. It’s not trying to land investment banking business by being too bullish on a particular company, but it can thwart confirmation bias.
“Even if we try to engage in proper research before making a decision, our minds will automatically pay more attention to information that supports our current beliefs. In this context, a useful thinking partner will help us acknowledge and consider different perspectives,” concludes Lamas.
