Facing that challenge has become an emerging and increasingly dense market, called interactive AI. Larger tech companies such as Microsoft, Amazon, Google and Oracle offer smaller companies and start-ups including Kore.ai, Omilia, Rasa, Senseforth.ai. , Verint and Yellow.ai.
Suppliers provide software tools that companies then customize and train on their own data.
This year, the business market for virtual assistants – aka chatbots – will grow 15 percent to more than $ 7 billion, according to Gartner’s forecast. Some of these boats are designed to help employees, but most are for customer service.
No company has traveled more politely and instructively than IBM for its chatbot technology. After defeating the human champion in his Watson Supercomputer TV game show “Jopardy!” About a decade ago, IBM set out to apply Watson’s natural language processing to other areas. Early focus was on the diagnosis and treatment of cancer, and IBM called health care a “moonshot.”
In January, after years of struggle, IBM announced that it was selling its Watson Health business to a private equity firm. A few days later, Gartner rated IBM’s Watson Assistant as the “leader” in AI for business conversations. Watson has gone from cancer moonshot to customer service chatbots.
Today Watson Assistant is a success story for IBM in the rest of its AI products, including software for data navigation and automation of business functions. The Watson Accessory has evolved over the years, constantly refining and improving. IBM quickly learned that the rigorous question-and-answer approach, while ideal for game shows, is very limited and complex in customer service settings.
“The real world has opened our eyes,” said Aya Sofer, vice president of AI technology at IBM Research.
The starting point for improvement, Dr. Sofer said there is a deep understanding of what happens in call-centers, working with other companies to mine and analyze thousands of calls between customers and human agents. In the dialogues, for example, tracking what questions and what follow-ups solve the customer’s problem, she said, and what were the so-called signs of a “bad conversation”.