Khoros has claimed that its patent-pending, next-gen "Khoros Customer Care Cloud" will address up to 80 percent of its customers' problems without a live agent in the loop.
In doing so, Khoros refers specifically to the virtual agent included within the Customer Care Cloud.
The generative AI-powered virtual agent – or “digital concierge”, as Khoros terms it – plugs into company knowledge sources to autonomously answer customer queries without prior training.
In utilizing the solution, businesses can build solutions within minutes, like the Klarna bot that – according to the financial services company – handles two-thirds of its live chat conversations.
Yet, with such a bot, businesses must employ a rigorous knowledge management strategy to ensure that the concierge delivers accurate, up-to-date information. Otherwise, their bots may dole out misleading responses and create many negative headlines – as New York City’s bot did in March.
Meanwhile, Air Canada recently got sued for its bot misinforming a customer and lost a court case arguing that the chatbot is a separate legal entity responsible for its own actions.
After these cases, Gartner warned contact centers against using AI to replace human agents – and other industry analysts have spoken out against autonomous GenAI use cases in customer service. The following LinkedIn post is a case in point.
With these warnings, businesses must proceed with caution when leveraging such GenAI-powered digital concierge models.
However, Khoros’s concierge goes further than many other models by extracting data from every customer interaction – driven by both live and virtual agents - into a “safe, self-learning system”.
That self-learning system forms new insights and generates support articles for the knowledge base – which the digital concierge feeds from, improving its outputs.
In this case, however, there is a human-in-the-loop who ensures the auto-generated knowledge article is high quality.




