[#VIRTUAL] KEYNOTE (AI): LivePerson -- Building a Mental Model Around Conversational AI: Why We Need to Teach How to Interact with Bots


Joe Bradley
LivePerson, Chief Scientist

Joe Bradley is Chief Scientist at LivePerson (Nasdaq: LPSN), a global leader in customer engagement solutions. LivePerson makes life easier for people and brands everywhere through trusted conversational AI and was recently named the #1 Most Innovative AI Company in the World by Fast Company. At LivePerson, Joe’s job is to combine conversational data and the skills of business professionals to teach computers how to talk to people and help solve problems.

Joe’s winding journey led him to business and AI through classical singing, English Literature (BA), public middle school teaching, Mathematics (MA), and Physics (PhD). This journey has been held together by a love of knowledge in many forms and the wide array of human perspectives that can create it. Joe is passionate about the intersection of science, technology, and human psychology, and believes that AI can (and should) be used to improve lives. Before joining LivePerson, Joe was a data science leader for Nike as it brought its eCommerce offering to life. Prior to Nike, Bradley was a data science leader at Amazon, working to personalize and simplify the shopping experiences of hundreds of millions of people. 


Use of conversational AI across retail, finance, healthcare, and other industries is on the rise. Whether they recognize it or not, today’s consumers are rapidly shifting their mindset — they are ready for, and even demand, a new type of interaction with brands centered around messaging: Indeed, new research shows that over 3 quarters (78%) of consumers want the ability to message with businesses and 83% would browse or buy products in messaging conversations.

Perhaps most importantly, consumers are suddenly, radically more open to automated conversations now than ever before: Positive sentiment towards chatbots nearly doubled in 2021 (61%) vs in 2020 (31%).

Despite new capabilities that make chatting with a conversational AI bot more like having a conversation with a human, there isn’t yet a prevailing mental model for what conversational AI is that will help people get the most out of their interactions with them. Simply put, people aren’t sure how to talk to bots. On the one hand, some people treat it like a search engine, typing in short commands; while others treat it like another human, telling long-winded stories and burying what question or issue it is they really are trying to address.

Similar to when search engines were first invented and people had to figure out how to effectively use them, many people may not know how to maximize the efficiency of a bot conversation. Tech companies can and must take the lead on that instruction to enable correct use of their products and to help users get the most benefit out of them.

During this session, Joe Bradley will offer guidance on how companies can help users find the middle ground of these two scenarios. How they can begin creating a playbook for cultivating best practices and interacting with conversational AI.

There are many questions around how companies should teach people to interact with conversational AI and how they can make this form of communication most successful that are just now being explored – How can we be sensitive to the fact that different people will respond to conversational AI in different ways? How can we help people learn and get the most out of this new type of interaction? Not only do these questions intersect with machine learning but they also involve psychology and sociology.

While few people have the time (or interest) in diving deep on how to best interact with conversational AI, bot builders can begin to offer clues and guidance on how to engage with conversational AI bots effectively. Having previously worked on data science and e-commerce projects at Amazon and Nike and advising brands like David’s Bridal and Virgin Atlantic at LivePerson on how to build their bot strategies, Joe Bradley will share his learnings on how to build a mental model around conversational AI that gets the most out of this increasingly used form of interaction.