Today, banks are dealing with an overwhelming morass of regulations – more than 20,000 new ones in 2015 alone. By the year 2020, there may be 300 million pages of banking regulations.
When I was in college, buying used textbooks had two benefits. It was a way to save a few dollars. Even better, if you could find one that was previously owned by a very good student, it may already have been accurately pre-highlighted, providing even stronger study assistance.
Wouldn’t it be nice if the specific regulations that applied to your business were highlighted? And maybe even sorted by country, by service, by deadline, by any number of variables you could select?
It’s not like banks aren’t throwing a lot of money at the problem. They’re spending about $100 billion a year on compliance efforts. And they’re paying more than $150 billion a year on fines and penalties when their efforts come up short because something was missed, misinterpreted or ignored.
The problem is that traditional technology is not well designed to understand and manage regulatory compliance. With regulatory guidance being completely unstructured (free form text in documents), it falls to people to painstakingly read and review countless documents in search of meaningful sections or passages. And as we know, human behavior, interpretation and concentration are variable.
Now, however, IBM is applying new technology to the field of regulatory compliance that can highlight critical information faster than any human. We call it cognitive compliance.
Cognitive compliance is based on cognitive computing, which gained prominence in 2011 when IBM’s Watson computer defeated the top human contestants on the televised quiz show, Jeopardy.
A cognitive computing system is based on four key principles:
- It interacts using natural language, context and reason
- It uses subject matter experts to collate and curate human intelligence, for rapid reuse or decision support
- It learns and improves with each new piece of information
- It builds speed and scale by using machine learning to handle complex, repetitive tasks
The technology behind Watson is showing real promise in the field of compliance.
To begin, in September of 2016, IBM announced it was acquiring Promontory Financial Group, a consultancy with a high percentage of former regulators. Now that the acquisition is completed, these subject-matter experts have already begun playing the critical role of training Watson in the science, art and language of regulatory compliance.
With the ability to ingest new regulations as they are created, and to review up to 800 million pages of text per second, Watson can provide decision support on compliance issues as diverse as regulatory reporting requirements for derivatives to consumer protection for mortgages. And a cognitive system can learn to alert the proper department if a new regulation applies to its business or mission.
In addition, proposed regulations also can be reviewed much earlier to ascertain their potential impacts on bank operations, allowing faster response time to engage in lobbying for or against a specific proposal.
Cognitive compliance can dramatically streamline the time and effort it takes to understand the potential obligations driven by constantly changing regulations, and enable banks to be better prepared to put in place the controls required to address those obligations. This will reduce risks and drive significant savings, which can enable banks to focus investments where it best serves any organization: driving client-focused innovation that increases business value.
That’s a highlight anyone can use.
