AI investment tools are not all created equal but global managers are looking opportunistically. The demand for AI investment assets has started an arms race between the largest banks and asset managers. Institutions are currently exploring “build versus buy” opportunities in the AI analytics space. Different adoption rates and operational friction exists but the industry is moving toward a more AI integrated investment environment.
Artificial intelligence (AI) in the investment world takes on all different shapes and sizes. More plainly stated, not all investment AI is equal. The EquBot team has intimate knowledge of bulge bracket banks and top tier asset managers initiating the “build versus buy” exercise specific to AI analytical assets. Behemoth fund operations with seemingly unlimited capital are favorably positioned to scoop up top talent in the machine learning and AI space but there are drawbacks inherent to the structures of these firms that allow for market disruption. The market development is not an unfamiliar story. Upon closer examination and extrapolating to another statistically heavy sport we are experiencing another “Moneyball” revolution albeit with AI on Wall Street.
The AI virtual arms race taking place within the banking and finance sector continues to heat up, and it carries the same enthusiasm as the drive to fill trading desks with quantitative researchers from decades ago. This development will promote a wide scale change but allows for disruption in the space through agile innovation and rapid deployment maneuvers from smaller players. This opportunity exists as a result of structural traditions and human bias. For these reasons EquBot was able to launch the world’s first AI powered ETF, and subsequent natural language processing NLP and sentiment related funds have surfaced from smaller investment operating firms. Leading investment firms are not easily willing to admit algorithms can handily beat their top portfolio management talent. Similarly, top portfolio managers are not willing to simply accept nor allow an algorithm to replace them in seats they have likely held for decades.
A side by side comparison of the two modes of operation – exclusively human versus robotic processing automation with AI, will highlight vast differences in capabilities. A human portfolio manager may speak a couple of languages and process a couple dozen articles a day during open market hours. AI portfolio platforms speak dozens of different languages, process millions of articles and financial statements, and operate around the clock. This simple comparison neglects experience, but an AI platform arguably can draw from hundreds of years of data with the appropriate training and will not be biased from human related factors such as relationships or ESG preferences. The summary comparison is quite compelling to leverage AI powered solutions within investment management operations to a high degree.
Given the longstanding careers of some of Wall Street’s finest, the integration of innovative AI technology will be muted and drawn out in large firms to avoid the “ruffling of any feathers”. The result is varying degrees of technological acceptance and it is understandable from both a firm and client perspective. Investors would question an immediate removal of a well known portfolio manager for an algorithm. A marketing campaign about the implementation of new trade technology would be more likely and accepted if introduced by a key portfolio manager. Ultimately, investor education levels are increasing and relationships have value, but data shows preference toward maximizing investment return. Given the reluctance to embrace a large scale AI transition at large investment managers, we expect to see increased innovation and market share capture from smaller AI investment focused asset management and technology operators.
From a cost perspective both traditional and AI powered operations have equivalent data costs, but the established firms carry key contributor risk as well as a higher operating costs inflated by rockstar portfolio manager and trader retention market rates. This underlies a critical theme. If human managers charging more for investment products begin to underperform AI competition that is also less costly from a fee and operating structure, the current framework will come into question.
This discussion highlights the AI Investment Moneyball moment. Smaller AI investment systems using sentiment, NLP, complete deep learning portfolio management modules are at the plate with scouting financial institutions looking to determine if there are opportunities. If these AI operations are left undisturbed and successfully outperform, we will see a significant shift in investment operational focus to the most innovative AI powered investment solutions.