Hindsight, as the idiom goes, is 20/20. Predicting the future is naturally far more difficult, but to those in the know, the foundations for the future state of market technology have already been laid.
In the first article in this issue of Waters celebrating the magazine’s 25th anniversary, we looked back over the events that have shaped markets over the past two decades, from the birth of electronic trading through to the terrible events of 9/11 and the financial crisis. In this article, Waters canvassed a range of senior figures from the worlds of investment banking, asset management, exchanges, technology providers and consultancy to give their views on what the formative technologies in the next phase of capital-markets evolution will be.
To regular readers of Waters, some of these topics will be familiar. Quantum computing, distributed ledger and artificial intelligence are all mentioned, of course, but for many, the real power of these emerging technologies lies not in isolation, but in combination.
Read on to discover what the future holds for trading floors, banks, hedge funds, and exchanges—and, indeed, society as a whole—from the people who are at the bleeding edge of developing tomorrow.
A singular evolution, and possibly, a singularity.
Quantum computing has emerged as one of the hottest technologies under development right now. Its advocates say that it has the potential to instigate fundamental changes in how markets operate—but also in how society as a whole evolves. In stark contrast to the hype machines in work around blockchain and similar technologies, however, even a cursory understanding of the technology suggests this may be more than mere hyperbole.
“We have a partnership now that’s been announced with IBM. In each line of business, we have a participant who’s looking and spotting opportunities in the finance world. I mean that’s further off than, I would say, blockchain, but it’s not that far off either. What that could do to our industry is incredible. What it can do to security and encryption, and everything else is incredible. I think we are big believers in that.”
Michael Urciuoli, CIO, JPMorgan Asset Management
“Of all the trends in fintech, the most promising one I think is quantum computing, which will increase the ability to due heavy-duty computing exponentially FASTER—bolded because this is big—and make practical some interesting algorithms. It will cause us to rebuild all our security and cryptography architectures as they will become hack-able, and will speed up adoption of all the other buzzwords: AI, deep learning, big data analytics, alt-data, etc. Our current cloud virtualization architectures will make this available to everyone, with all the major tech players investing in this technology.”
Keith Lubell, CTO, Berkery Noyes
“It’s hard to pick any one technology as being the answer to the world’s problems and that’s true here, as well. Quantum is a super exciting space. I’m an applied physicist by training and in school—undergrad and graduate—I remember well solving textbook theoretical problems in quantum mechanics and it was fun, in and of itself, because it had a unique set of rules and challenges, but I truly didn’t believe—foresee—that we’d have workable quantum computers in my lifetime available to run experiments and to really see this be applied to real-world problems. That is incredibly exciting.”
Kathryn Guarini, vice president, IBM Research
The machine comes to life, in more ways than were imagined.
The laziest trick in journalism is to use a Skynet metaphor—or any quote from The Terminator films—to introduce a section on artificial intelligence. It’s also ridiculous—artificial intelligence, as a discipline, incorporates any number of subsets, from machine learning through to deep learning and the use of big data. The emergence of available compute power coupled with data growth has, of course, breathed new life into theories that are over 30 years old at this point, but for perhaps the first time since the AI Winter, there’s a palpable sense that this will truly change how we view the fusion of man and machine.
“Our exchanges have now processed a record 52 million transactions in a single day. That massive volume of data presents both opportunities and challenges that can be managed with deep learning and data science technologies. Neural networks can help CME and our clients discover new market opportunities. A decade ago, it would have been inconceivable for traders to have readily-available access to daily changes in crop health, oil storage capacity or base metal smelting activity. This is now both possible and advancing, as evident in recent CME Group collaborations with Orbital Insight, RS Metrics, and Tellus Labs. Simultaneously these technologies and their capacity for analysis enable us to continually advance how we monitor our markets and ensure integrity. These intersecting technology trends are key to the future of capital markets, and we look forward to being at the forefront as they develop.”
Bryan Durkin, president, CME Group.
“While the near-perfect humanoids of the hit HBO TV show Westworld may always be fantasy, the convergence of human intelligence and artificial intelligence in the investment process—from idea generation to trading—is rapidly approaching. The hand-wringing and consternation on the threat of automation to existing roles tends to block out the enormous opportunity to transform the active management investment process. The opportunity will not be achieved merely through the proliferation of alternative data and the rapid advancements in artificial intelligence; it will be a creative wave of portfolio managers, traders, and risk managers who will reinvent active management. It is not about letting purely systematic strategies take over the industry, but rather adapting the investment and trading process to a rapidly changing global economy.”
Adam Sussman, head of market structure and liquidity partnerships, Liquidnet.
“Artificial Intelligence has the potential to be the most impactful advancement of our time as we entrust more of our personal day-to-day lives to technology. Today, we produce and store an enormous amount of data; couple that with the compute power at scale now available to us, and our insights become limitless. It will be an exponential change in how we apply technology’s predictive and interactive capabilities. AI done well creates job opportunity. The pace of change, however, is happening at such a rate that it has the potential to fragment societal structure—social, human and legal, to name a few. As with all innovation, AI runs the risk of outpacing our ability to understand and act on its implications. We need to gain a foothold of being ahead by working towards a universally-accepted foundation for its use. Changes in the workforce are predictable, and we are accountable to be thinking about the impact of technology and transforming the workforce so we are prepared. Responsible AI is not about what we can do; it is about what we should do.”
Cathy Bessant, COO and CTO, Bank of America Merrill Lynch
“There is no doubt to me that the automation of more/most-existing workflows using basic technologies and eventually machine learning will continue to drive major efficiency in all financial functions. It will accelerate significantly from here and those that embrace and harness the power of data and machine learning will thrive and the rest will be left behind.”
William Murphy, CTO, Blackstone Group
“Artificial intelligence technologies are already being used today much more than people think. That being said, I think we have just begun to scratch the surface of the opportunities for these technologies, and in the next decade, the impact will be profound. Machine learning, natural language processing/natural language generation, visual recognition and satellite-based applications will all provide the opportunities for companies to create new products, find and service clients, and improve operational efficiency in new and exciting ways. Fasten your seatbelt!”
Michael Radziemski, former CIO, Lord Abbett & Co.
The future of all these technologies lies in their fusion.
The benefits of cloud computing have been well documented over the past 10 years, not least of all in the pages of Waters magazine. For many, the use of private cloud structures are a well-ingrained form of technology, but the shift to the use of public cloud, and the emergence of secure, institutional-grade providers such as Amazon, Microsoft, IBM, Google, and others have intertwined with the development of other technologies to provide a foundation layer for the future of market technology. As Kathryn Guarini, vice president at IBM Research says: “Any one technology by itself can’t do all the things we need it to do to drive technology innovation—it’s the combination.”
“I think the capabilities of the future are quantum in a hybrid—and it’s going to be delivered via cloud. I think cloud just changes a lot of things that we can’t even predict yet. And then quantum computing will be made available by the leading cloud providers. So it’s almost like specialized compute, just like your GPU is, and now TPUs with tensor processing units, you’re going to be able to add quantum where necessary for simulation and for new applications that we can’t even envision yet.”
Brad Peterson, CIO and CTO, Nasdaq
“Some of the biggest areas we’re looking at as enabling change over the long term, are cloud computing, machine learning, and AI, and shifts in how clients think about build versus buy or outsourcing to specialist third parties. This “best of breed” mentality is extending beyond infrastructure and into processes as well, like regulatory compliance and monitoring, performance attribution, and risk modeling. Clients are turning to firms that specialize in areas such as these, stay current with the latest developments, and offer real-time monitoring and tools that empower practitioners to step in where needed.”
Rob Robie, head of analytics, FactSet.
“I think quantum computing will have a massive impact, but still, that would help and support and drive the way we use machine learning in this industry. So, I would say the whole buildout of machine learning, the combination of machines and humans in the way we make decisions and improve things, that I think will be the big game changer in this industry. But in a way, we also have to remember what the enabler for that will be. That is basically a cloud build up, and there is basically a bunch of other stuff that needs to be there in order for machine learning to be the big transformer of this industry. So, if we can take for granted that we will have a broader transition to cloud, we will get cheaper compute, yes, then I think machine learning is the one to pick.”
Magnus Haglind, head of product management, Market Technology, Nasdaq.
“I think quantum computing will be the most important computer technology for the 21st Century. Any particular problem, however, whether it’s in fintech or anything else, is going to be a combination of everything else. Every single place you look you’ll see a combination of classical methods, new AI methods, and quantum. So it’s another great edition to our toolbox and it opens up this amazing intellectual frontier: People get to think about problems in ways they never could before, and we know there are so many really smart people out there who will take advantage of it.”
Bob Sutor, vice president of IBM Q Strategy and Ecosystem, IBM Research
But What About…?
Weaving together other threads into a cohesive technology whole.
Predicting the future, as mentioned at the start of this article, is a difficult business. Outside of a few common threads, those also interviewed by Waters had a number of other theories about where important developments would take place.
“I believe it will have a seismic and formative impact on the industry. Specifically the ethereum branch for smart contracts. In the short term (5 to 10 years) I think structured products and derivatives will all be managed via distributed-ledger tech. This will result in incredible back-office efficiencies and translate directly into market growth via transparency, reduced barriers to entry, operational efficiency, et cetera.”
Henry Kravchenko, head of IT, Marshall Wace
“The adoption of blockchain technology, which already is a proven tool for accessing and trading certain assets, is going to continue to revolutionize our financial systems. I think we’re going to see it applied to more mainstream transactions, where it will be attractive because of its intellectual property and privacy advantages. It’s going to be exciting to watch.”
Seth Thomson, CIO, DRW
The Internet of Things
“Mobile transactions (including banking) become more prevalent further eliminating the need to cash, FX exchange, credit cards, smartphones, etc. by taking advantage of technologies such as face and voice recognition. Sensors will be everywhere as will the software to interpret events picked up by these sensors. This will facilitate billing by usage which will change the nature of insurance risk and encourage ride sharing, bike sharing, and the general sharing and trading of small quantities of goods and services. True robo investment decisions will become prevalent as neurological tests are able to measure the emotional biases of investors to implement fully customized, automated strategies.”
Elliot Noma, managing director, Garrett Asset Management
Voice Recognition, Augmented Reality and Virtual Reality
“Technology is going to start to know what’s happening, where you’re heading and where you might need to be heading, and start to serve it up for you in advance. To that point, it’s not going to be one technology, it will be a lot of technologies working together in unison. But think about Siri and Alexa and I see a day where there’s going to be something like one of those devices in every office and in every conference room and you’ll just ask, ‘What where the earnings last quarter? What are the projected earnings this quarter?’ It’s going to be all voice-enabled where you’ll just talk to the device in the room to get all the information you want.”
Richard Alexander, managing director, IT, Cerberus Capital Management.
“Large financial institutions are already heavily investing in cybersecurity defenses and are secure—but are they secure enough? Innovative technology and machine-based learning, such as AI, can help safeguard financial institutions against cyber threats even further. Intelligent systems can do everything from monitoring all network activity to analyzing raw data to identify patterns—all in much quicker time than a human could if they were processing the information manually. In fact, powerful AI engines—such as IBM Watson—can go as far as detecting and flagging suspicious activity and in the near future, it may also devise defense recommendations, too. Financial institutions need to deploy these first-class technologies in order to stay on top of, and ahead of, the changing threats within the cyber-security landscape. Proactive prevention is essential if finance organizations want to protect their data.”
Thomas Zeeb, head of securities and exchanges, SIX.
“While much of the buzz (and hope) around the next 25 years may currently be focused on blockchain and quantum computing for financial technology, I see the devolution of the operating system as the most significant paradigm shift on the horizon for the financial technology industry. While many productivity applications continue to rely on Windows for user interactions, and development of next-generation applications orient themselves on mobile platforms, some have pondered the real race for an operating-system-agnostic approach is being waged in browsers. But in reality, the next generation of the user interface has yet to truly show itself, I expect in the next 25 years the eventual solution will take elements of virtual reality and augmented reality along with data visualization functionality as the underpinning to the coming great user challenges we’re facing with larger datasets and more complex applications performing more and more sophisticated workflow and business functions. I can hardly wait to see what evolves and how the various industry forces interact as solutions develop.”
Charles Walters, CIO, Halcyon Capital Management.
Bryan Cross, who heads UBS Asset Management's QED group, joins to discuss alternative data and AI.Subscribe to Weekly Wrap emails
- The Problem Solver: Paul Bari, Nordea
- JP Morgan's FX Algo Tool Launches on Bloomberg Terminal
- Refinitiv Consolidates Data Insights on China's Belt and Road Initiative
- Wavelength Podcast Episode 153: Nasdaq's Lars Ottersgård
- CAT’s Tale: How Thesys, the SROs and the SEC Mishandled the Consolidated Audit Trail