Scroll Top

The world’s largest hedge fund is using AI to automate its entire management team

article_hedgefunddao

WHY THIS MATTERS IN BRIEF

The news is filled with news about automation and machines that are taking our jobs, arguably for the first time we can see companies who are automating both the workers at the base of the pyramid, and the executives at the top. In the next ten years, or perhaps even sooner, we could see the first fully automated company that doesn’t employ anyone.

 

Recently I wrote an article about Foxconn, the world’s largest manufacturing company, whose management team have just announced they are going to automate away the jobs of over 1.2 million workers, but now we have a company that’s busy flipping this on its head and trying to automate its entire management team. And, Bridgewater Associates LP, like Foxconn, is no toy town company – it’s the titan in its industry and the world’s largest hedge fund with over $160 Billion of assets under management.

Furthermore, if you put these two companies programs together you get something that begins to look very similar to what futurists and venture capitalists call a Distributed Autonomous Organisation (DAO) – a company that runs all by itself, without the need for any humans, or human intervention.

I encourage you to think about this for ten seconds – I know you’re busy and thanks – think where this could take us. Could you imagine a “DAO Uber”? There’s the ultimate DAO “startup” right there, after all, they’re already automating their transportation people and processes – automating the rest of the company, from the customer service staff to the finance teams, would be a comparative breeze. Or, could you imagine one of Uber’s competitors getting the jump on them using a DAO model? After all a DOA would, technically at least, cost peanuts to operate.

The future is coming, whether you see it or not – it’s just a case of joining the dots, and it’s time to get armed and get savvy before you’re left wondering what happened to your career, or why your company just got blindsided by disruption.

 

See also
DeepMinds AI based WaveNet tech makes computers sound human

 

Anyway, I digress – back to Bridgewater.

Working deep inside Bridgewater is a small army of software engineers busy at work on a secret project that founder Ray Dalio calls “Principles Operating System” (PriOS).

Driven by the mantra “meaningful work, and meaningful relationships” Dalio, who is known for his accurate macroeconomic predictions, founded Bridgewater in 1975 from his two bed Manhattan apartment, and he’s committed to creating a company built on trust, transparency and technology – although many have questioned his vision, and his methodology.

His latest goal is to build a platform that will fully automate all of the firm’s managers and management – the people and the processes – and if it’s successful the program will represent the culmination of Dalio’s life work – to build Bridgewater into an altar to radical openness – and a place that can endure without him. Although again, when we talk about openness, as you’ll see, humping all of a company’s management into an artificial intelligence (AI) system that could quickly become it’s own “Black Box” – albeit one that one day might be able to explain all of it’s actions – might run counter to that goal. But for now the company is driving forwards with its plans.

Under Dalio’s gaze most of the company’s meetings are recorded, employees are expected to criticise one another openly – in order to help each other improve, people are subject to frequent probes of their weaknesses, and personal performance is assessed on a host of data points. It sounds arcane, and 35% of new hires leave within the first eighteen months, unable to cope, or some would say, adapt to, the company’s unique culture. However, those who get over the initial shock to the system stay for ten years or more. It’s a marmite company – either you love it or you hate it, there doesn’t seem to be much middle ground.

Bridgewater’s new PriOS system will enshrine this rather unorthodox management approach into a software system that’ll be able to tell staff how they should spend every aspect of their days, even down to whether or not they should make a particular phone call, and while the platform is still under wraps one employee familiar with it described it as “like trying to make Ray’s brain into a computer.” Bearing in mind that, according to Institutional Investor’s Alpha, Dalio earned $1.4 Billion last year and that Bridgewater made twice as much profit as any of its rivals last year, maybe that’s not a bad thing…

The rules for Bridgewater’s staff are laid out in a 123 page public manifesto known as “Principles” which every employee is expected to know and diligently apply and it contains snappy maxims such as “By and large, you will get what you deserve over time,” and “Don’t ‘pick your battles.’ Fight them all.”

 

See also
Scientists use “Predator particles” to create super mice with infra red vision

 

Dalio, who originally stepped back to take on a mentoring role six years ago, stepped back into the CEO’s seat in 2016 and a few weeks later he gathered the managers together and said the firm had grown bloated and inefficient. The fix, he said, would be a “renovation,” where weak employees would be let go. Staff cuts began almost immediately. Since his return, head count is down by about 150, or 10%, with potentially hundreds more on their way out.

 

The underpinning to his success, Dalio often says, is his belief that markets reflect the workings of a misunderstood economic machine, and interpreting its mechanics requires a relentless and often painful dedication to getting to the truth through “thoughtful disagreement.” That’s why employees are encouraged to challenge each other repeatedly and without reservation.

The economy, Dalio has written, is “really just a zillion simple things working together,” and decades before computer driven trading came into vogue, Bridgewater began tracking relations among hundreds of millions of separate data points – such as international interest rates and retail sales – and created investment algorithms. The main hedge fund embodying these algorithms, Pure Alpha, uses the data to buy and sell stocks, bonds, currencies and other assets, and the fund has long anticipated booms and busts around the world, including the 2008 global financial crisis which it predicted as early as 2006.

Dalio also believes humans work like machines – a word that appears 84 times in the Principles – and it’s this thinking that’s prompted him to create PriOS. The problem, he has often said, is that people are prevented from achieving their best performance by emotional interference. It is something he thinks can be overcome through systematic practice and that applies to managing, too. Successful managers “design a ‘machine’ consisting of the right people doing the right things to get what they want,” he wrote in the Principles.

 

See also
This new test will tell us if AI has become self-aware and gained consciousness

 

PriOS, therefore, is an attempt to make management nearly as systematic as the firm’s investing process.

As you might expect data on the firm’s staff is incorporated into PriOS from a multitude of sources – ranging from personality tests and exams to peer review and performance and Dalio is always in search of new data with which to measure his staff. He once raised the idea of using head bands to assess people’s brain waves, according to one former employee, but unlike some banks, the idea wasn’t adopted.

The push by Bridgewater to automate management addresses a larger challenge, which is how this culture would survive without Dalio, who is 67 years old.

When he temporarily retreated from day to day management in 2010, Dalio spent weeks at a time away from the firm’s Westport campus indulging pursuits such as scuba diving and, until he gave it up, bow hunting. And to run Bridgewater in his absence, he brought in a collection of big name outsiders, but before long he was expressing frustration with some of them.

There was James Comey, who Dalio hired as general counsel in 2010, saying he would act as a “godfather” figure to improve justice at the firm – within three years, according to two former employees, Dalio had taken to calling Comey something else, a “chirper” who repeats stale ideas, as opposed to a “shaper” – the Bridgewater ideal of a visionary leader. Comey left in 2013, telling colleagues his personality didn’t fit, and now you might better know him as none other than the director of the FBI, and another executive, long seen as Dalio’s heir apparent, Greg Jensen, ran afoul of his boss a year ago for allegedly talking about him beyond his back, which at Bridgewater is a sin second only to dishonesty. In a tribunal the firm convened, Jensen was found to have broken the rules and he lost his co-CEO title but remains co-chief investment officer.

 

See also
New brain reading computer lets paralysed patients communicate by thought

 

At the core of PriOS is a walled off engineering group called the Systematized Intelligence Lab, headed by David Ferrucci and some of the team who led the development of the artificial intelligence (AI) system IBM Watson, before joining Bridgewater in 2013.

Although outsiders expected Ferrucci would use his talents to help find hidden signals in the financial markets his job has focused more narrowly on working to analyse the torrent of data the firm gathers about its employees which includes ratings employees give each other throughout the work day, called “dots.”

The Systematized Intelligence Lab has been involved in helping to design several mobile apps that form part of employees’ everyday lives, and among them is one called the “Dot Collector.” It’s pretty much as it sounds – it collects data points, or “dots,” on employees. The apps do a range of things – everything from letting employees rate each other on dozens of attributes, as well as holding snap polls on issues during meetings – including asking blunt questions such as whether a current conversation is a waste of time. And all of these millions of dots are collected, analysed, and blended together to produce “Baseball Cards” that show people’s strengths and weaknesses in various categories, such as “touching the nerve” – a prized attribute.

A job advertisement for the Systematized Intelligence Lab said it sought to “extract meaning and domain understanding from the wealth of text-based data our innovative applications generate,” and several employees said they have felt that working at Bridgewater was as much experimental research into human decision making as it was investing.

Recently new apps have helped to shed more light on Dalio’s expanding technological vision. Software called “The Contract,” loaded onto staff iPads, instructs employees to formalize the goals to be achieved over time and tracks how reliably they follow through. Meanwhile, an app called “The Coach” lets people input a question and directs them to the relevant passage in Dalio’s Principles document. The goal is to evolve The Coach into an intelligent system that can assist in decision making – and those are just the initial uses of PriOS.

In the future use cases will also include the ability to scan open positions at the company and have PriOS sort through the staff to find people with particular talents and strengths to fill them, and in other instances, employees at loggerheads over decisions won’t have to hash out each debate out loud – they would either key their opinions into PriOS, or talk to it using natural language, and it’ll rank their perspectives, consult with Dalio’s Principles, and spit out the best way to proceed.

 

See also
Experts are starting to agree that AI will replace CEO's

 

The ultimate vision is that PriOS would be able to predict outcomes of meetings before they are completed, and to guide people to take certain actions throughout the day, and by 2021, Dalio wants at least three quarters of all management decisions to be made and communicated by PriOS, and he envisions that most of the remaining people at the firm would be there, not to take individual choices, but to fine tune the system and intervene when something isn’t working.

Dalio’s original manifesto nods to the goal of automating decision making, “with greater use of the Principles,” it says, “not only will they be understood, but they will evolve from ‘Ray’s principles’ to ‘our principles’ and Ray will fade out of the picture.”

Related Posts

Leave a comment

FREE! DOWNLOAD THE 2024 EMERGING TECHNOLOGY AND TRENDS CODEXES!DOWNLOAD

Awesome! You're now subscribed.

Pin It on Pinterest

Share This