PE Tech Report


Like this article?

Sign up to our free newsletter

SEI identifies five key trends in disruptive technology – No1: Watsonisation

In the first of a five-part series, SEI discusses the impact of five disruptive technologies and how they can be applied to the financial industry. The first of these five trends relates to IBM Watson, and the power of cognitive computing. 

Cognitive computing and machine learning capabilities are becoming so great that even Elon Musk has raised concerns. 

Speaking recently at Recode’s Code Conference, he said: “I don’t know a lot of people who love the idea of living under a despot,” the inference being that self-awareness super-computers could become a malign force; think creepy-voiced HAL9000 in the seminal movie, 2001:  A Space Odyssey. 

Such are the advances in computing power under Moore’s Law that businesses of all shapes and persuasions – not just the asset management industry – are leveraging cognitive computing systems that are becoming smarter and using human speech to carry out instructions; SIRI in the iPhone for example. 

Machine algorithms are imbuing cognitive computing systems and represent the next stage of artificial intelligence; one that harnesses predictive analytic capabilities to not just improve tactical activities such as client service, but overall enhance the way that we live our lives. These are systems that can simulate human thought, while learning as they continuously consume and process vast amounts of data. 

Recently, when thinking about the implications of technology, SEI identified five unfolding areas of innovation that are reshaping our world. The result was a white paper entitled “The upside of disruption – why the future of asset management depends on innovation”. 

In the paper, those five trends are referred to as:

  • Watsonisation
  • Googlisation 
  • Amazonisation 
  • Uberisation 
  • Twitterisation


The following section will focus on the first of these trends, ‘Watsonisation’.

This term comes from the IBM Watson supercomputer developed by the company’s founder, Thomas J. Watson. As has been well documented, Watson shocked the world in 2011 when it won the game show Jeopardy!. It provided a snapshot of the future; a world where cognitive computers become more adept at mimicking human thought processes.

“At the turn of the millennium, the business of asset management was far less complex that it is today. Tailwinds were everywhere and the financial crisis wasn’t even a glint in anyone’s eyes.  Yet today, with headwinds more omnipresent, in order for fund managers to continue to be successful, they also need to be operationally adept and take advantage of new fintech advances,” says Ross Ellis (pictured), Vice President and Managing Director of the Knowledge Partnership in the Investment Manager Services division at SEI. 

“But it’s not just about seeing what our peers are doing and new technology within the asset management industry, however. We need to look at what’s happening in other industries and what seemingly non-competitive firms are doing there, such as IBM with Watson, Uber, AirBnB and Twitter. It behooves us as industry participants to look at every possibility in the global marketplace to see where we can take advantage of technology and deliver better solutions to our constituents,” he adds.

Open architecture

What Watson has done has opened up a portal into a new technological dimension. According to Fast Company, in 2014 more than 5,000 applications developers were working in the Watson ecosystem alone. 

“IBM has invited entrepreneurs, clients and prospective customers to see how the Watson platform can transform their businesses.  They have created an open architecture environment in such a way that encourages participation by a large number of developers. By building a foundation that allows people to come in and develop unique applications, it changes how people interact with and consume data. Indeed, dozens of APIs now exist in the platform, many of which would never have occurred had it not been for the open environment.  That is the beauty of what IBM is doing with Watson,” says Jim Warren, Head of Solutions Strategy & Development at SEI.  

The whole premise of Watson and the potential of cognitive computing is pushing people to develop solutions in all kinds of unique areas; wearable devices for the medical industry, GPS technologies (Uber), facial recognition systems to identify card counters, and even analysing the tone in people’s writing. 

“Because it is an open architecture, there are no limits to how it can be used. It opens up so many opportunities,” says Ellis. 

One could argue that the silicon chip has heralded in the second Industrial Revolution; the Technology Revolution. Asset managers such as Bridgewater Associates have been quick to recognise the power of cognitive computing. It now has a dedicated artificial intelligence unit headed up by former IBM executive, David Ferrucci. The unit is responsible for developing trading algorithms that make predictions based on historical data and learn how to respond to changing markets. 

A self-direction tool

This is, in effect, incorporating a robo-advisor component to the firm; but rather than advise investors, the algorithm suite uses predictive capabilities to improve Bridgewater’s trading expertise. Indeed, in a recent talk on innovation and competitiveness at Harvard Law School, SEC Commissioner Kara M. Stein cited four developments that are challenging the financial services industry’s existing business models and practices: robo-advisors, online peer-to peer lending, equity crowdfunding, and blockchain technology.

“We would cite robo-advisors as being one example of the Watsonisation trend,” says Ellis. “By using algorithms and technology to do asset allocation quickly, it allows participants to partake in the investment realm in a more straightforward way. Whether or not someone uses this alongside a personal advisor, it’s a tool that provides self-direction that previously couldn’t have been done as inexpensively, efficiently, or accurately. It also allows people who have previously never participated in investments to do so.”

Robo-advisor tools have been used in the funds industry for quite some time; they just weren’t referred to as such. Rather they were used to produce model portfolios for HNW investors at private banks. Where the disruption element is coming to play is that today’s robo-advisors are forcing change in the industry because they are now allowing retail investors – and those who have never participated in the investment process before – to do something that was previously the preserve of the wealthy. 

Google DeepMind

Warren says that one of the ways that SEI is trying to grow its business is to use technology for lower value, albeit important, tasks in places where humans used to do it; i.e. taking the robot out of the human, and using cognitive computing to perform mundane tasks. “This is being done in areas such as managing exceptions, for example. We can use predictive analytics to manage (future) exceptions by spotting them before they even happen.”

Machine learning reached another milestone recently when Google DeepMind, the tech giant’s artificial intelligence division that builds self-learning algorithms, achieved a 4-1 series win with AlphaGo over Go world champion, Lee Se-dol. Using neural networks for deep learning, AlphaGo studied 30 million positions and played simultaneously with 50 different computers to learn the game. 

In another example, using machine learning, Google DeepMind is being used to cut energy costs in Google’s data centres by up to 40%.  

To further demonstrate the power of cognitive computing, in the real estate industry it is allowing asset owners to monitor the performance of buildings in real time and use predictive analytics to detect when boilers or metering systems might be about to break down and need replacing. 

Such is the breadth of capabilities with Watsonisation that the biggest challenge for fund managers is knowing how and where to best embrace new technologies to drive efficiency and spot new insights. 

How far can AI really go?

“One way to look at this technology is how it can reduce the amount of input that operational teams need to do. At SEI, we are consuming, aggregating and presenting more information, more quickly and more comprehensively, on behalf of our clients. Many of our clients leverage us from a platform perspective; and by doing so they are able to make smarter, more insightful decisions based on the quality and quantity of data we have accumulated. 

“Clients are starting to embrace this concept of, ‘How do I cut down on time wasted on analysis that isn’t necessary?’  And, ‘How can I spend more of my time of higher value activities?’

“The fascinating thing is that two years from now, the conversation will likely be vastly different because so much more would have happened. Our perspective is that we need to stay on top of all the technologies and maintain an open environment because I don’t think any of us know how far AI capabilities can go,” says Warren. 

Ellis expands on this point, saying that SEI is looking to figure out what its clients are seeking and provide it to them the next time proactively “so that they don’t have to go through multiple steps to get what they ultimately want.”  In essence, SEI can react in a way that pre-empts what their clients want to do in the same way that Google’s search engine pre-empts what we want to search for.  

“There are some things that machines, or technology, can now do that makes the jobs that we do of higher value and higher quality – it also frees people up to more engaging work, and improves the customer experience,” he concludes.

This is how AI is able to turn out information on data that already exists to better predict what will happen next. Take cybersecurity. There are now algorithms that will predict where a hacker is likely going to be before they have a chance to do anything and put firewalls in place to counteract the threat before it happens.

“The technology now is such that it can be customised to every single person, not large groups; with robo-advisors, for example, you can customise your exact requirements to create a portfolio that meets your specific needs,” says Warren. 

Whilst AI might be useful in overcoming cognitive biases in stock picking, when it comes to active management there is still an argument that a human overlay is needed to add an additional level of trust, provide someone to discuss ideas, or even override an AI system when necessary. To date, robo-advisors appear to be more suited for passive management but as technology improves, it could well give active portfolio managers a run for their money.  

“Technology is additive. You can use Watsonisation to streamline things and add efficiencies, but in our business, having a connection to people, and looking them in the eye, is important. It’s about marrying technologies to benefit your clients, and improving the solutions provided, while at the same time retaining a human connection with those clients,” concludes Warren.  

To read the SEI white paper in full, please click on the following link: 


Like this article? Sign up to our free newsletter