Shep Hyken\'s Customer Service Blog

Guest Blog: Assistive AI for Real Work

This week we feature an article by Christopher Connolly who shares fascinating information about enterprise-grade assistive AI and how it can help make customer support more effective.  AI has come a long way in recognizing the content – and context – of customers’ requests and questions. – Shep Hyken

Enterprise-grade assistive AI initiatives are already in evidence across various industries. There’s been plenty of publicity around IBM’s Watson and precision medicine, but that’s not the only game in town. According to The Verge, “Microsoft’s Project Hanover aims to help doctors narrow their search for the right drug regime, sifting through medical papers to suggest the most effective treatments. IBM is working on a similar project, using 600,000 medical and 1.5 million patient records to explore better cancer treatments.”

These systems don’t have to target only life and death issues. In July, Google launched a People + AI Research (PAIR) program to “democratize” AI, focusing on its helpfulness to humans, and exploring how AI systems can “amplify the expertise of doctors, technicians, designers, farmers, musicians, and more.” Microsoft launched a similar effort the same week. This concept recognizes the larger truth about our expanding AI capabilities. In essence, this technology has the potential to help people in nearly any endeavor. At Genesys, we make customer experience management systems that use assistive AI development — we call it “blended AI” and have personified it as “Kate.” Kate aims to augment the capabilities of call center agents and customer service/helpdesk personnel. It is a hybrid execution leveraging automation and machine learning so real agents can better handle customer inquiries. That’s it. The technical implementation combines a collection of technologies — chatbots, micro-applications, knowledge center, widgets, a dialog engine, engagement engine, APIs, etc. — under an integration layer. This assists human agents with customer interaction information, sentiment analysis, suggestions on how to respond to queries, help to execute processes and providing guidance on what to do next during an interaction.

This blended AI model doesn’t eliminate the need for customer service reps. But it can significantly help a select group of people perform specific duties with greater ease and acuity — and that’s the whole point of enterprise-grade assistive AI. Applied to a target task, it can serve a purpose beyond the mere novelty displayed in consumer AI personas, but it does not replace the person. As noted by MIT Technology Review’s Jamie Condliffe, “The implication is that AI shouldn’t necessarily take over from humans but, rather, help them do a better a job.”

Christopher Connolly is the vice president of solution strategy for Genesys, the global leader in omnichannel customer experience and contact center solutions. He is based in the company’s regional offices in Miami and can be reached at

For more articles from Shep Hyken and his guest contributors go to

Read Shep’s latest Forbes Article: Walmart Moves From Traditional Retailer To High-Tech Innovator


Leave a Reply