How conversational AI can remove sensitive information from contact center calls

Every day, people call customer contact centers and voice sensitive information, such as credit card numbers, to agents. Now, a conversational artificial intelligence (AI) solution using natural language understanding capabilities offers a way to remove this information from calls, while still transmitting data for transactions.

This is important because processing any type of personally identifiable information (PII) inevitably involves a range of compliance with security and privacy regulations which may vary depending on the jurisdiction. There is also a significant risk that sensitive information could potentially be leaked or stolen. In fact, there are known incidents where voice-provided credit card information has been written down by malicious agents, leading to unwanted results.

“There was an incident where a corporate client came to us with a true story saying, look, this happened, someone wrote down the credit card numbers and these things got leaked on the free market", Srini Bangalore. , head of AI research at conversational AI provider Interactions, told VentureBeat. "It got us to start thinking about the technology itself and how to remove personally identifiable information in real time with low latency, without affecting the user experience."

To this end, Interactions has developed a new technology, Trustera, which is generally available today. The goal is to use artificial intelligence and machine learning (ML) techniques to identify PII in real time, remove it from the live voice call, and pass the information to the underlying digital systems for transactions in an encrypted approach.

Take a Hybrid AI Approach to Conversational AI

Interactions is a company that designs conversational AI technology platforms for organizations.

Conversational AI technology is commonly associated with human interactions with bots, but this is not the approach that Interactions has widely taken. Bangalore said his company took what he called a hybrid AI approach.

With the hybrid AI model, humans are part of the process alongside conversational AI to help support the user experience in a frictionless approach. The Trustera system, for example, is not driven by robots, but is intended to work in environments where a person calls a customer support center and then speaks with a human.

Bangalore said the process of removing personal information in human-led conversations is more complicated than for purely robotic and digital interactions in an interactive voice response (IVR)-like system. He noted that in IVR or bot conversations, the system knows when PIIs are transmitted because they are part of the process and are initiated by the system.

With human-directed conversations, it's not always at the same point in a conversation that personal information is requested or transferred. It is also necessary to understand what personal information is being sent, as well as to understand the actual human speaker.

How Trustera's conversational AI works to secure PII

The AI ​​technology that Interactions has developed for its conversational AI platforms has its roots in capabilities from AT&T Bell Laboratories.

In 2014, Interactions acquired speech analytics technologies from AT&T, where Bangalore had previously worked for 18 years. Speech recognition capabilities have steadily improved over the ensuing years, with the integration of Natural Language Understanding (NLU) functionality, which enables the Trustera service.

Interactions trained its guard information model to understand when different human speakers transfer personal information. The model is not static and is constantly updated.

"We have a self-supervised automatic ML approach, where we take calls from the day before and we have a theoretical confidence metric to say that these are data items that we can add back to the model," said Bengaluru. "So we update the model periodically that way as well."

VentureBeat's mission is to be a digital marketplace for technical decision makers to learn about transformative enterprise technologies and transact business. Discover our Briefings.