As financial markets become increasingly complex, many institutions adopt new technologies – especially AI – to support their investment process. The combination of human intelligence and a rising maturity in AI is gradually building a novel approach to building uncertainty-proof portfolios, refining strategic and tactical asset allocation, and launching modern investment solutions. Besides, the rise of Open Innovation partnerships and strategic partnerships with Fintech companies is fueling a modern approach to investing and launching new investment solutions, making a case for AI adoption a priority for many financial institutions to keep, or seize, a new market share and increase revenues.
At the beginning of 2004, LEGO’s revenues were down by 40%. Market research suggested that it was due to the rising popularity of video games amongst younger generations. Until then, the company relied exclusively on Big Data to guide its strategy, indicating that millennials were impatient and more impulsive than the generation before them. Accordingly, LEGO executives decided to reduce toy complexity by having fewer pieces in bigger sizes. However, this move was unable to stop the decline. Simpler toys were easy to sell but less fun and quickly forgotten. The turning point occurred when the Big Data met the ‘Small Data’ collected with 1-to-1 interviews conducted by the companies’ executives. It turned out that adults and children saw LEGOs as a challenge to win or an enigma to solve. The company strategy pivoted: smaller pieces, more complex and realistic constructions (like the Taj Mahal, 5922 pieces), and partnerships to involve a newly digitally influenced audience. Ten years passed, and revenues exceeded two billion dollars, making LEGO the primary toy producer in the world.
The LEGO anecdote should push every executive towards a thought-provoking reflection: excessive simplification is investment risk. Unreliable approximations or information gaps harm decision-making. Only new and original knowledge can help better inform investment decisions. Big Data alone won’t do the job. Rich Data will combine a rigorous approach to data analysis with human intelligence and expertise.
To a certain degree, the fast technological growth has allowed the opportunity to extend how far investment managers can build their market view. In other words, good technology creates a frictionless conversion of data into decisions.
Yet, before any new technologies become helpful – especially in a financial context, where mistakes are costly – they can take years, if not decades. Those who witnessed the transition of Machine Learning from Data Science to Econometrics departments know this well. Bill Gates once said, “We always overestimate the change that will occur in the next two years and underestimate the change that will occur in the next ten.” This is the case for Neural Networks for Asset Allocation, which took over 40 years to become investment managers’ tools to build adaptive portfolios.
To this point, the annual Gartner Maturity Curve measures the level of maturity acquired by a specific technology. In AI, the most mature practice is currently taken by ‘predictive analytics’ – instruments and models that learn to comprehend the context, suggest alternatives, and support decisions. In practical terms, AI is helping investment managers to answer questions like:
According to a Deloitte study, financial institutions leading the adoption of Artificial Intelligence in the investment process had positive repercussions on several KPIs such as AuM, productivity, revenue, and market share. Recent innovation (in products and processes) demonstrated how numerous activities can be carried out more effectively. That makes winning or losing no longer a question of embracing the new but rather learning how to use it correctly.
Today, an investment manager’s real challenge is distinguishing the relevant information in data. Human minds are not made to process data the same way as a computer that is, on the contrary, designed to reason to make a data-driven decision. In this light, integrating AI’s investment views within investment processes means that managers can count on an additional and effective tool that can sift through data and provide the relevant information to guide their investment decision-making.
AI has been successfully adopted in various company processes. A recent study by Stanford, along with Google, Bloomberg, and McKinsey, showed the adoption rate of new technologies amongst large institutions, including institutional investors. The report indicates that processes, risk management, and product innovation are the areas in which AI is most heavily adopted. Thanks to the automation of low value-added activities, these areas have received a boost from AI adoption, since it allows human resources to participate in tasks of greater value, thus significantly increasing performance and revenues.
The transition towards a new regime in financial markets, combined with the rise of traditional (and novel) risk factors is pushing Asset and Wealth Managers to permanently redefine how investment decisions are made. In this context, AI-Driven PredictiveAnalytics, in particular, have become widely adopted and mature to support managers in their investment decision-making process – for example, to update and refine Strategic and Tactical Asset Allocation. This also pushes institutions to collaborate in a spirit of Open Innovation to remain competitive and shorten the time-to-market of new investment solutions, such as AI-driven discretionary mandates. The case for adopting AI to support the investment process also makes sense from a business perspective as new research points out how financial institutions that have adopted AIgenerate more value at the enterprise level, thus creating a virtuous cycle between performance and revenues.
Originally published on Finextra.com