The robotic manager
Ray Dalio, the founder of the world’s largest hedge fund, has been working for some time now on creating artificial intelligence able to comprehensively manage his company. And by managing, we mean managing; this is, including decisions on investments, hiring and firing employees, or even deciding when to make phone calls to customers.
Bridgewater Associates manages $165 billion. Its founder is searching for ways to reduce the impact of human variability, implying that managers are to all intents and purposes imperfect systems influenced by difficult or impossible to control factors such as mood, perceptions, etc., that lead them to make decisions that could be improved. The company has been working on a series of algorithms capable of learning from all the decisions made in the company, examining them according to the results obtained, as well as evaluating all the people involved in them through a system that assigns each of them a performance rating.
The unit, called Systematized Intelligence Lab, is led by David Ferrucci, the principal researcher who helped IBM win Jeopardy in 2011. In the case of Bridgewater, the algorithms feed on data generated by a company that collects absolutely everything, including recording and processing all meetings, along with an ongoing evaluation of each team member’s performance by their cohort. If everything goes according to plan, within five years, three quarters of corporate decisions will be made in an automated manner, taking into account the general principles and the vision that the founder has for the company. Basically, as one of the linked articles explains, the idea is to try to replicate the brain of the founder through artificial intelligence.
This is a subject we discussed about a month ago: automation no longer applies solely to mechanical, repetitive tasks, the so-called three Ds (Dull, Dirty and Dangerous). There are few better examples of a white-collar company than Bridgewater, investing real money, hiring highly qualified people and providing them with sufficient resources to work for a company where the majority of decisions are made by artificial intelligence, albeit under the strategic guidelines and objectives established by a person.
What will happen if this company outperforms traditionally managed hedge funds, where human beings make the day-to-day decisions? in an environment where competitors are constantly compared according to objective and well-established parameters, how long would it take, before we saw a race to adopt this type of technology?
(En español, aquí)
