Conversation As Interface

Conversation as Interface

By Kolby Harvey, PhD, Chief of Conversation


One of the things we as a company are most excited about is conversational AI’s ability to realize a decades-old vision of computing: asking the computer a question, just like you would a person, and receiving an answer in return, just like you would from a person, without coding, making a spreadsheet, or googling. Now, it’s as simple as saying your device’s wake word and firing off questions. At the heart of this leap forward is conversation, not just in a given machine’s capacity to parse language, but in a complete reimagining of computer interfaces, where conversation itself is the interface and bots, in the words of author Amir Shevat, add a new layer to “an already established human interaction.” When interacting with voice systems, the spoken word becomes the tool that we employ to get what we want, manipulating speech rather than the graphical elements of a computer screen. It’s important to keep in mind that without conversation, none of it works. Even the most robust AI code is useless if the conversational interface that houses it isn’t up to snuff. In a 2016 article published in IT Professional, Seth Earley breaks it down: “Humans create knowledge, while machines process, store, and act on that knowledge.” For Earley, AI is nothing more than “applied human knowledge.” When we verbally interact with our voice assistants, Alexa/Siri/etc. and the answers they return are the application of human knowledge programmed by developers, which are then accessed/retrieved through conversation.

As AI implementations increase across industries, organizations must think critically about how to overlay conversational interfaces on top of existing systems. What will we need to reimagine and what can be carried over? Think about the transition from terminal user interfaces to GUIs: while some of the hidden/less visible aspects of computers may have been preserved, the front-end experience of using a computer changed tremendously. With each new application of conversational AI systems, new methods of engagement must be devised. Remember that AI “isn’t magic,” and is instead the result of human knowledge and action. It requires, in Earley’s words, “foundational structures that can be reused across many different processes, departments, and applications.” Simply put, it’s not a fresh coat of a paint on the family home; it’s a partial (possibly more) renovation.

While it may require time and hard work, the benefits will be worth it in the end. Earley writes that the “true power” of AI will be realized when it’s placed in a holistic framework of machine-intelligence-enabled infrastructure,” arguing that organizations which invest in “product and content architecture, customer data, and analytics, and the harmonization of tools in the customer engagement ecosystem” will “gain significant advantage over the competition.”

At InfinitAI, we are of the same mind as Earley. Our Digital Contact Center and Agent Assist products are designed to reinvigorate existing structures while integrating seamlessly with them. We see the Digital Contact Center and Agent Assist as the first steps on a longer journey of reimagining how organizations and their customers interact with and use AI systems. We look forward to pushing the limits of what conversational interfaces can do, how they are designed, and what we use them for.



  1. Designing Bots: Creating Conversational Experiences by Amir Shevat
  2. “There is no AI Without IA” by Seth Earley



Kolby Harvey, PhD
Chief of Conversation

Kolby is a writer, designer, and artist living in Washington state. In 2018, he earned the University of Colorado’s first creative doctorate in Intermedia Art, Writing and Performance.

Add Comment