Interview with David Shrier, Founder & CEO, Distilled Analytics

Tell us about your journey into Artificial Intelligence? What made you start an AI-driven financial services company?

I’ve been fascinated with computers since I was a child. I grew up on science fiction like Isaac Asimov and William Gibson that extrapolated the impacts, benefits, and dangers of AI. While building my career in financial services and data analytics, I grew really interested in the concept that, as we get more and more big data, and computers get smarter and smarter, human/machine hybrid systems will be able to predict future events and improve financial decision-making in ways that science fiction writers barely dreamed of. My co-founder, Sandy Pentland, and I are developing a book on this topic called “Extended Intelligence” at the MIT Media Lab. I launched Distilled Analytics in January 2017 and, together with the top minds in both academia and industry, we are pioneering a new model of behavioral analytics to transform financial services.

Define your ‘Ideal Customer’ profile?

Our ideal customers are people and organizations who need to grow their businesses, better serve their customers, and have more impact with their capital. Within that, we seek people who aren’t trapped into using inadequate, last-generation solutions, but rather are prepared to be seen as innovators and leaders in their fields.

What are the core tenets of your business strategy focusing on the Fintech industry?

We want to deliver better insights leading to better decisions and provide our clients with heretofore unprecedented foresight and transparency by leveraging machine learning and big data.

In 2018-2020, what are the biggest challenges for financial institutions? How could technology solve these challenges?

Adoption of AI, advanced analytics, and blockchain will be accelerating over the next two to three years. This will cause dramatic disruption as technology companies like Amazon, Google, Alibaba, and Tencent go head-to-head with conventional financial institutions. Converging tech innovations means rapid shift in market share as consumers and businesses alike gravitate toward the best user experience at the lowest cost. We will also see new markets open up. For example, in our estimation more than $10 trillion of credit risk is mispriced due to reliance on FICO-derived bureau models that rely on linear regression. Our techniques leveraging behavioral analytics can provide both a better understanding of the risk of the existing “banked” market, and the ability to reliably lend to the underbanked and unbanked markets by way of novel data insights driven by AI-powered systems.

How do you build new methods of behavioral data science applied to financial models?

We seek to understand human behavior and predict what will happen in the future based on this behavior. We have found our models to be 30-50% more predictive than many existing models when deployed against real world data from actual bank customers. We calibrate our array of analytic models against a specific context in bank or asset management, as appropriate, and then provide tools to our customers to support their ability to make better decisions.

How do you make AI deliver economic benefits as well as social goodwill?

One of the fascinating areas where we are applying behavioral analytics is in impact investing. We estimate that more than $5 Trillion of capital is being deployed into investments with both profit and purpose. Investors are seeking a market rate-or-better return on their investments, and also want to have positive impact on the world through their capital. A MIT Sloan market assessment determined that 89% of ultra-high net worth investors are seeking investments with profit and purpose, but only 43% of their money managers feel this is important. When the professional money managers were asked why, they indicated a frustration with how to measure the nonfinancial factors associated with an investment. We know how to measure return, but how to we measure impact?

Distilled Analytics applies new research out of MIT to provide investors transparency into what is happening outside of simple economic returns. This includes decoding the human factors associated with a financial investment to assess job creation, crime reduction, even poverty alleviation via enhancement to micro GDP. It’s a powerful application of AI to investments, and could have transformational impact on the world as we mobilize ultimately trillions of dollars of capital towards intelligently investing with profit and purpose (more than $60 trillion of asset managers have signed on to the UN Principles for Responsible Investment, illustrating the demand for impact).

What are your predictions for the AI market in 2018?

AI will continue to expand into broader and broader usage not only in the consumer financial services markets, but also in more institutional applications. What’s interesting is that the first wave of $18 Billion of Fintech investments went largely (73%) into consumer financial applications, and only 4% went into investment banking. Prospectively, we anticipate more and more investment applied to institutional financial services applications, which opens up $65 Trillion of equity and debt, and over $200 trillion of real estate assets.

What is your vision in making AI technology more accessible to local financial communities? Do you provide any teaching or learning programs for banking and non-financial identities?

We are actively seeking to help improve capabilities of financial executives around AI and other disruptive technologies. In my teaching work, my OxfordFintech.org program is an immersive “fast forward” button for working professionals to understand and apply financial innovation around new tech like AI.

What AI startups and labs are you keenly following?

The Vector Institute in Toronto is a fascinating project, albeit for a very specific school of thought. They’ve attracted a good deal of attention. The most exciting work, in my belief, is going on among several labs at MIT, particularly the MIT Media Lab, as we look to create “centaurs” that are part-human, part-AI, and deliver superior insights and results to either human or AI alone. This “Extended Intelligence” is the topic of our next book that I mentioned, which compiles research of some of the leading academic and industry players exploring this important area.

What technologies within AI and computing are you interested in?

I have personally become fascinated with the ethics of AI, and how we can apply Responsible Innovation principles when adopting new technology. The auto industry in the US in the 1980s and 1990s, and likewise the heavy industries in the Midlands in the UK, raced to adopt automation without taking into account the human cost of throwing millions of people out of work without reskilling them. We’ve created a generation of permanently-unemployed and angry voters as a consequence, and are now reaping the costs of that narrow-sighted action. With new generation AI, we may do the same to millions of bankers and even the 5% of global employment that’s associated with the trucking industry.

What would happen if we took some of those savings, as Accenture is doing, and reinvested them into reskilling and retraining our workforce to better adapt to this new future? This is Responsible Innovation. We saw this theme resonating at Davos this year, and expect to see it continue to be discussed throughout 2018 and beyond.

As an AI leader, what industries you think would be fastest to adopting AI/ML with smooth efficiency? What are the new emerging markets for AI technology markets?

Other than financial services, there are a lot of interesting things happening in healthcare using AI and ML. Autonomous vehicles hold the potential to relieve traffic jams and save lives, perhaps offsetting some of the damage they will do to trucking employment. Home automation has been adopted with lightning speed, although I note the dangers of inviting poorly-secured listening devices into our homes. More work needs to be done around how we adopt and implement these technologies so that we don’t create further unintended consequences.

What’s your smartest work related shortcut or productivity hack?

Finding great people! With a great team you can get almost anything done. Sometimes, it can take a bit of time to find them and get them set-up but then you multiply your efficiency by order of magnitude more than with automation. Innovation is a team-driven function, and designing teams of great people creates a “social intelligence” that can take on large-scale problems.

Tag the one person in the industry whose answers to these questions you would love to read:

Tan Le at EMOTIV

Thank you, Scott! That was fun and hope to see you back on AiThority soon.