Threat Management

Meet ViViAN, a new ID crime chatbot that may be used for future B2B cyber applications

Chatbot ViViAN icon (ITRC)

A new AI-based chatbot tool used to help identity crime victims seek after-hours help was also designed with future B2B applications in mind, including helping employees report a cyberattack when the IT or security team is unavailable.

Meet ViViAN, short for Virtual Victim Assistance Network. This chatbot helper is a new service currently undergoing beta testing by the Identity Theft Resource Center (ITRC), leveraging technology developed by its partner SAS Institute. Thanks to ViViAN, individuals do not have to wait until normal ITRC business hours in order to report an incident; rather, they can lodge their complaints with the chatbot and receive reassurance and guidance on the immediate next steps they should take.

All communications with ViViAN are then later followed up by a live agent when one becomes available. But at least this way, victims are able to act swiftly when their data is at stake and time is of the essence.

“We run a contact center that has a defined schedule. And we’re on the West Coast, so that defined schedule is not always the most convenient for people, particularly on the East Coast and [during] overnights, weekends, holidays,” said ITRC Chief Operating Officer James Lee, in an interview with SC Media. “Identity crimes aren’t committed on a set schedule. So we have always been looking for ways to be able to provide more service to victims when they need it.”

And same principles behind this tool can also be applied in business situations should ViViAN in the future be converted into a commercial solution. “ViViAN has to be able to determine what issue is most relevant to the person and what is the best advice to give them – accurately and quickly,” said a joint written statement from Lee and ITRC CEO Eva Velasquez. “B2B and even government agencies could use this kind of chatbot in complex circumstances to help diagnose issues before working with an advisor or to offer solutions. Today, those processes more often than not involve a customer/employee/consumer wandering around a website looking for the right information which they may or may not ever find.”

That’s especially true for “small businesses… that don’t have the same level of IT support” as larger organizations, said Lee during his separate interview with SC Media. Such companies could use the tool for various help desk and incident response needs: “My Outlook isn't working – what do I do? I think I've been phished – what do I do? I clicked on a link I shouldn't have… All those kinds of things could easily be handled by a system like this,” he explained.

The tool could even apply to business functions outside of cybersecurity such as inventory support – anything that requires a rich, jargon-heavy business lexicon or taxonomy.

Indeed, that’s the unique aspect to this solution. This chatbot’s natural language processing abilities (including functionality for understanding, conversation flow and language generation) were designed to address topics and issues that are typically more linguistically complicated to describe than the average chatbot request. In other words: the “chat equivalent of a Boolean search,” through which someone asks a complex question and hopefully “you get a more refined response back in the process,” said Lee.

“Most chatbots that you see in a commercial world, they have a very defined process and they have a very defined vocabulary,” said Lee. “When I'm ordering a pizza, there's only… so many ways I describe it, and most people know how to describe the pizza. That's not true at all when you get into a very complex subject like identity crimes. People don't know what terms to use, they don't oftentimes even know how to begin to describe it. Wo we had to spend time working on the defining the language that people use.”

To accomplish that, the ITRC fed into the chatbot’s AI engine years of anonymized transcripts of conversations between its live advisors and ID theft victims. “We have taken the most common situations people initially start with, and the myriad number of ways that they described that [and] we’ve gotten it refined to a point where now ViViAN can recognize that,” he said.

As needed, the chatbot will ask additional questions to clarify the predicament in which the victims find themselves and then make suggestions to remediate the issue, whether it’s, for example, a stolen driver’s license or someone fraudulently filing for unemployment under someone else's name.

It was also necessary to make ViViAN sound sympathetic to the plight of the victim, as opposed to an emotionless robot.

“That’s more or less sort a hallmark of the ITRC. We have always… treated identity crimes holistically. It's more than just the financial impact on people.” The impact is emotional too, he explained, and according to the ITRC, victims can process such experiences similarly to victims of violent crimes.

“So when we started looking at the language and the tone and the pace [of ViViAN], that was very much top of mind… That the language that was used, the tone that it reflects, was empathetic and knowledgeable, and not in any way judgmental.”

Jared Peterson, director of advanced analytics at SAS, said another reason that ViViAN is ripe for B2B usage is because the tools used to build it, SAS Conversation Designer and SAS Viya, can accept SAS analytics and reporting into its chat flow, allowing employees to inquire such data-driven questions as “How many customer complaints have been filed on product ‘X’ in the last 24 hours?," “How many employees answered our latest survey?” or “Does this purchase have the hallmarks of a legitimate or fraudulent transaction?”

“One of our primary goals was making it easy for chatbot designers to build conversational interfaces to their analytics,” said Peterson. “It seemed natural to connect these two things – conversations with chatbots and analytics/models for answering questions. We wanted to build a chatbot framework that would make it straightforward to access those capabilities and insights from within a bot, to make it easy for chatbot designers to access the SAS platform.”

SAS presented details on ViViAN at its virtual 2021 SAS Global Forum last month.

The ITRC envisions scenarios where a version of ViViAN could be adopted by not only businesses, but other victim and human services organizations, consumer education programs, nonprofits and government organizations. 

The ITRC and SAS Institute received federal funding to create and develop ViViAN after submitting a proposal three years ago to a grant program offered by the Department of Justice and its Office for Victims of Crime. “It was all about trying to find new ways to serve victims, particularly as technology becomes more and more embedded in people's lives,” said Lee.

Bradley Barth

As director of multimedia content strategy at CyberRisk Alliance, Bradley Barth develops content for online conferences, webcasts, podcasts video/multimedia projects — often serving as moderator or host. For nearly six years, he wrote and reported for SC Media as deputy editor and, before that, senior reporter. He was previously a program executive with the tech-focused PR firm Voxus. Past journalistic experience includes stints as business editor at Executive Technology, a staff writer at New York Sportscene and a freelance journalist covering travel and entertainment. In his spare time, Bradley also writes screenplays.

Get daily email updates

SC Media's daily must-read of the most current and pressing daily news

By clicking the Subscribe button below, you agree to SC Media Terms and Conditions and Privacy Policy.