Bitcoin Money 20/20 Panel: Artificial Intelligence and Machine Learning
Ryan Strauss | Oct 31, 2016 | 12:00
On Wednesday morning at the Money20/20 conference in Las Vegas, a unique and timely panel discussed the emergence and potential applications of Artificial Intelligence (AI) and Machine Learning (ML) in the global, traditional payments arena.
Also read: Technical Analysis: Bitcoin Price to Hit $1500 By Year-End?
Bringing Blockchain to Artificial Intelligence and Machine Learning
As computational technology advances, leveraging trends in data and meta-data will help organizations understand both their customers and other businesses more extensively. AI and ML are going to affect all realms of society, and payments are not immune to this trend.
Democratization of tools for analytics in the field will help open up doors for an expanded crowd. As tools and APIs for developers looking towards AI or ML expand, developers will be able to access these complex tools more easily.
Speaking to this, Dr. Arif Ahmed of U.S. Bank remarked how, “With deep learning, you have better ways to conceptualize problems. You see how voice recognition, fraud recognition, and more are improving. You start with the technology, and then you host concepts. . .We have reached a point where this will exponentially increase over the next few years.”
Pattern recognition from AI and ML advancements will have a strong impact as it relates to Anti-Money-Laundering (AML) and Know-Your-Customer (KYC) practices. Particularly in the litigation response matters, Husayn Kassai of Onfido explained how often times remediation work today is outdated.
“The current way that it is carried out isn’t necessarily fit for a digital age,” said Kassai. “It doesn’t make sense to have fully human authentication systems at a bank.”
Ensuring a proper intake of data will be key here, the panel said, as financial services players transition to updated or increasingly distributed backend platforms.
In the future, many consumer-facing products, including chatbots, will make their way into digital services. For lots of financial players, the ability of machines to understand human slander falls short, as placing consumer-facing concerns in context is a major challenge.
People can build chatbots with specific purposes, such as manuals to build a plane or figure out the nature of a mortgage contracts. To minimize errors, look for chatbots in financial services to be developed with specific purposes, such as mortgage loan contracts or ATM interfacing.
David Gilvin of IBM remarked how “AI is always on, 365, 24/7. . .If the machine is making the decision, then fundamentally it is not the same as the financial advisor, it is automated through machine learning. . .not only is it always on, it is everywhere.”
Due to increased regulatory pressures and oversight, AI development in financial services is in a stranglehold. Banks and traditional financial institutions will look towards AI once business models emerge for ways to profit from solution-grade AI-financial software.
Blockchain technology, however, presents a more emergent structure for data storage and application processing such that its inclusion in traditional financial institution circles could lend more easily to AI and ML applications.
Inevitably, the panel said, the regulatory environments at home and abroad will have to adopt. Speaking to this, Martina King of Featurespace echoed the challenges that banks faced in the early days of the internet, and how primitive regulation frameworks created during this timeframe can be changed to help AI and ML applications in banking along.
Finally, in the future, improvements will become more clear to everyone. Stakes for taking an early lead in proprietary software markets in AI and ML to financial services institutions signals the large amounts of money currently on the table.
Adding additional programmatic layers on top of existing disparate financial data should yield massive insights for supply chain providers, retailers, customers, and businesses worldwide. Look for both AI and ML to move away from “black boxes” and more towards proofs-of-concept, similar to the trend that is ongoing in the permissioned-blockchain world. Both applications will likely have a massive impact in the near future!
What do you think about the possibility of Artificial Intelligence or Machine Learning as it applies to financial services, or even blockchain technology? Share your thoughts in the comments below!
Images courtesy of Ryan Strauss.
