Chatbots are a buzz-worthy topic for all organizations, including credit unions. Members can interact with chatbots in the digital channels they use regularly, like messaging apps and smart speakers, to receive personalized service and support. Chatbots can help fulfill common tasks like verifying account balances, paying bills, and transferring funds between accounts without having to speak with a member service representative. Implementing a chatbot can provide a custom-tailored experience for members while streamlining internal operations for credit unions.
Although chatbots are designed to be user-friendly and intuitive, understanding how they work can be a challenge. There’s complex technology running behind the scenes that make chatbot interactions so effortless. You may have heard terms like artificial intelligence, natural language processing, and machine learning in relationship to chatbots. What do these terms mean and how do they come together to create a chatbot? Here’s a shortlist of chatbot-related terminology to help get you started.
- CHATBOT – A chatbot is a computer program that’s designed to simulate a two-way conversation with another human. Chatbots rely on text input (for example, a user typing a question and submitting it to a chatbot via Facebook Messenger) or speech input (for example, a user initiating a chatbot conversation with voice commands via a smart speaker like Amazon Alexa). For credit unions, chatbots can be an automated means of interacting with members to answer questions and deliver service around the clock.
- VIRTUAL ASSISTANT – A chatbot might seem similar to a virtual assistant, but they are two different technologies. Virtual assistants (Apple’s Siri, for example) perform simple task completion like checking the weather or locating the nearest restaurant. Although virtual assistants make use of some aspects of artificial intelligence, they aren’t a chatbot because they’re focused on task completion rather than two-way conversation. Another key distinction is that chatbots can chain together several different instructions to achieve goals while virtual assistants typically do not.
- ARTIFICIAL INTELLIGENCE (AI) – AI allows a computer program to complete tasks that would otherwise require human intervention. For example, facial recognition once required both cameras and human beings to make positive identifications. Computer programs now perform this task quickly and at scale by digitizing facial characteristics to find matches. Another example is language translation. Translation programs powered by AI can now translate speech in real time to break down barriers in interpersonal communication.
- MACHINE LEARNING (ML) – ML is a subset of AI that allows computer systems to learn and improve from experience without direct programming. The technology allows programs to learn from what they do successfully as well as from the mistakes they make, giving them the ability to grow without explicit intervention from a programmer.
- NATURAL LANGUAGE PROCESSING (NLP) – NLP is a subset of computer science that’s focused on the interaction between computer programs and human languages. NLP combines with ML and AI to understand commands submitted by a human in their own language. Using bill payments as an example, a member might ask a similar question in many ways. For instance, “What do I owe this month?” “What is my current bill?” “How much is my balance?” “What do I need to pay?” These questions are all worded differently, but the intention behind them is the same. NLP helps computer systems understand that, although each question is unique, all should yield the same answer.