There is no “the” solution for complex phenomena such as AI or humans.
It’s difficult not to compare AI to algorithmic trading strategies which were, and still often are, viewed as “the solution” to making profits in the market. Algorithmic traders now manage approximately one-third of the assets of the world’s $3 trillion hedge fund industry, including models that input data like a company’s profitability, volatility, or interest rates shifts to make trading decisions. Many have produced spectacular returns and pride themselves on preventing human emotion from clouding their trading judgment. However, when market surprises occur, not only are these strategies no longer effective but they can end up resulting in a total loss. Recently, due to long periods of significant underperformance investors, who once saw this strategy as a failsafe, are now leaving these trading vehicles in droves.
Currently, the way we think about problem-solving is ineffective. Having one strategy to rule them all is not going to be effective in a world filled with complexity. One solution to market trading, business, sports, people, or any other complex phenomenon is not enough to deal with them effectively. We need all possible strategies at our disposal and when we pick one as the answer we kill off the possibility of all others. As soon as we turn something that once worked in a particular moment into the answer, it becomes the proverbial hammer and everything we see is a nail. Our strategies become a crutch and we become convinced we need them to be effective.
As the algorithmic trading strategies in the above example illustrate, the “the answer” dogma has become problematic. Instead of leaning solely on one tool, we can relate to them as one of many possible and valuable tools. This keeping of everything open to us, rather than closing down the possibility of other solutions, is of paramount importance when it comes to being effective in a world of complexity.