Pega’s Intelligent Virtual Assistant Makes Chatbots Smarter

Chatbots are promising, but many of them simply don't add business value from a customer service perspective—yet. It will take time before the technology matures enough to handle complicated customer service tasks, but with the introduction today of the Pega Intelligence Virtual Assistant, Pegasystems is setting out to accelerate the process by offering artificial intelligence extenders for platforms including Facebook, Slack, and others.

"For channels like Facebook, Alexa, and others, they control how much of the richness they enable when it comes to customer service interactions. With the Pega Intelligent Virtual Assistant, we're adding a layer of intelligence that [might otherwise not exist]," says Ying Chen, head of product strategy for the office of the CTO at Pegasystems.

Pega Intelligent Virtual Assistant adds Pega's enterprise application to existing support platforms to give them more advanced capabilities. Facebook Messenger’s existing chatbot technology, for example, doesn't use advanced AI techniques such as deep learning, relying instead on a more basic artificial intelligence platform, Wi.ai, which Facebook acquired in 2015. Plus, the Facebook Messenger bot is a channel-specific solution that does not always communicate seamlessly with other channels.

"Pega's virtual assistant for Facebook however, uses natural language processing to deliver smarter, more helpful conversations," Chen says. "We're also differentiating ourselves [from the Messenger bot] by implementing machine learning to provide a more natural cognitive bridge between bot and customer."

The virtual assistant layer can extend the capabilities of other types of bots as well, including Amazon Alexa and Google Home. Users can also build extenders for additional platforms themselves in a code-free environment, including WeChat and Skype. 

To improve bot functionality, Pega's Intelligent Virtual Assistant leverages predictive analytics and decisioning to provide the right solution for customers and uses natural language processing alongside text analytics to improve engagements with bots. And though the virtual assistant is built to improve interactions with bots, it can also facilitate the smooth transition between bot and human communication.

"There's a real disconnect now in customer service interactions. Our goal is to take advantage of artificial intelligence as well as a conversational user interface but go beyond the one-dimensional silos. If the conversation needs to shift from a chatbot to a person, it should be seamless and pick up where it left off," Chen says.

The Pega Intelligence Virtual Assistant is available as part of the Pega Platform.

Related Articles

The Pega Platform just got an update to its customer service application, including a number of new artificial intelligence features aimed at shortening how long it takes agents to resolve requests, cut down on errors, and boost customer satisfaction while freeing up agents' time to work on complex requests.

Posted October 25, 2017

A lack of transparency with AI solutions can be problematic, but Pegasystems' T-Switch aims to mitigate the issue by giving organizations more control over their AI engagements.

Posted September 29, 2017

Customer engagement platform provider Pegasystems has added artificial intelligence-powered capabilities to its CRM solution, Pega CRM. The new tools are built to improve customer service agent performance by pulling insight from desktop solutions that agents rely on and identifying any processes that slow down daily activities.

Posted March 03, 2017

Pegasystems today announced its acquisition of OpenSpan, a provider of robotic process automation (RPA) and workforce analytics software, bringing together robotics, analytics, and case management to enable more productive contact center employees.

Posted April 12, 2016