Watch all the Transform 2020 sessions on-demand here.
Conversable is rolling out a new system today for its bot creation platform that allows clients to visually edit their own machine learning in order to better answer their customers’ questions.
Since its launch last June, Conversable has worked with customers like Budweiser and made bots for TGI Friday’s, Sam’s Club, and Whole Foods.
Called Answering Questions Using AI (AQUA), the machine learning system is the brainchild of Conversable cofounder Andrew Busey, creator of iChat.
The company believes it has solved what cofounder Ben Lamm describes as “a more advanced message-based FAQ management problem.”
June 5th: The AI Audit in NYC
Join us next week in NYC to engage with top executive leaders, delving into strategies for auditing AI models to ensure fairness, optimal performance, and ethical compliance across diverse organizations. Secure your attendance for this exclusive invite-only event.
People ask bots a lot of questions, and the visual interface found on the Conversable bot platform is made to help companies answer customers’ typical and most frequently asked questions, as well as less frequent or one-off questions. The platform draws in chat questions from about a dozen platforms like Twitter, Facebook Messenger, Amazon’s Alexa or Google Home.
“How do we present FAQ where it’s not just a page of listed questions? How do we gain insights and train the system…? How do we use the collective knowledge of other inquiries coming in and deploy to various messaging platforms?” Lamm said in a phone interview with VentureBeat.
The machine learning trainer was made to give people with little understanding of coding languages or more complicated systems the ability to train machine learning-powered bots. Conversable also tracks how effective people are at training the system.
“We’re starting to look at some gamification elements to make internal training of the system a little fun and competitive,” Lamm said.
Not every company wants or needs complicated forms of artificial intelligence, Lamm told VentureBeat.
“The big data deep learning things are great and they’re incredibly well-funded, but a lot of companies — even publicly traded companies — don’t want to focus there. So you have to start with how do you help companies build on a conversational model and structured data?” he said.
The AI to train FAQs is the start, Lamm said. More visual interfaces to train artificial intelligence are on the way.
“There’s a couple other ideas that we have so that we’ll have multiple views that are going to be writing to the Conversable platform over time. That will effectively do different things for different people,” Lamm said.
