The State of Conversational Artificial Intelligence
Ever since viewers watched the crew of Star Trek’s U.S.S. Enterprise talking with the ship’s computer, using natural language to execute queries, conduct research or control the ship, typing on a keyboard has seemed like an antiquated stepping stone on the road to something better.
With the introduction of Apple’s Siri, Amazon’s Alexa, Microsoft’s Cortana and Google Assistant, conversational artificial intelligence (AI) has finally made its way to the masses. All over the world, people are talking and interacting with their phones, tablets and virtual assistants. Whether it be checking email, conducting a search, booking a flight, calling an Uber or just passing the time by listening to corny jokes, people have become accustomed to, and comfortable with, the reality of interacting with AI.
Conversational AI has already begun transforming many industries, and projections show the trend will only continue. In fact, Gartner’s research shows that “by 2020, customers will manage 85% of their relationship with the enterprise without interacting with a human.”
Given the rapid progress of this new technology—projected to grow to more than $11 billion by 2023—2019 promises to be a significant year in the development of conversational AI. Let’s look at some of the trends that will impact conversational AI in the year ahead.
As mentioned, conversational AI will increasingly play a major role in how customers interact with the enterprise. Similar to chatbots before them, conversational AI engines will enable companies to cut costs.
One key area where the technology will differ from its predecessor is the intent behind its use. When chatbots first appeared, many companies saw them as a way to slash customer service staff and rely almost solely on automation to assist customers. The reality, however, is that a significant portion of customers still want to talk to a human being at some point in the sales and/or service experience.
Conversational AI offers the best of both worlds. By providing a natural style of communication that people are accustomed to, the technology will increasingly be used to filter and sort customers’ service requests. Because of the advancement over traditional chatbots, many customers may be happy to use conversational AI systems for easy-to-solve issues. For more complex problems, the technology will serve as an important first step, gathering information, asking counter-questions and directing customers to the human agent best equipped to help them.
Thanks to the AI aspect of the technology, each interaction will lead to more intelligent and efficient responses, lowering cost even more while simultaneously improving customer satisfaction.
One of the biggest ways conversational AI will change industries is in the realm of search. For decades, search has largely relied on keyword-driven parameters to dig down and help the end-user find what they’re looking for. Every savvy searcher has memorized the necessary methods and keywords to include specific terms, omit vague connectives and drill down to the results they want.
Conversational search, in contrast, will use AI to make the process much more natural and productive from the beginning of the search. Rather than typing archaic keywords, a user will be able to ask questions, using natural language, and have the AI respond. In the case of easy queries, the AI may respond with a clear answer immediately. In more complex scenarios, the AI may ask open-ended questions to clarify the user’s intent, similar to the back-and-forth nature of a conversation between two humans, essentially duplicating the Star Trek-style experience referenced above.
While this advanced search capability won’t see widespread adoption in 2019, this year will see it make major inroads in specific industries, such as banking, retail and automotive websites. With their relatively narrow focus, these industries will offer the perfect testbed for conversational AI, as opposed to traditional search engines where users perform an infinite variety of searches.
As conversational AI continues to evolve and become more intelligent, the possibilities to control and automate daily tasks will increase. Amazon’s Alexa can already be used to play music, make online purchases and more. As the technology evolves, however, AI will allow people to control more of their surroundings simply through conversation.
For example, a person may come home and tell the dishwasher to run, ask the house if anyone knocked on the door while they were away and tell the TV to display a favorite show. The individual components are already in place: smart doorbells, virtual assistants and smart TVs and appliances. In 2019, however, we will start to see a convergence of these technologies, all of which will be accessible via conversational AI.
As the technology continues to grow and improve, one overriding concern will be the safety and integrity of the data being collected. For AI to function properly, it needs gargantuan quantities of data to analyze, pattern and use. Even an AI’s ability to grow and improve its efficiency depends on continued data input.
This will represent one of the greatest challenges for conversational AI in 2019 and beyond. With major corporations suffering data breaches, others—such as Facebook—misusing customer data and increasingly strict legislation on the part of the European Union and other governments, finding a way for AI systems to gain, store, use and protect the massive amount of personal data they have access to will be the single most important hurdle to overcome.
2019 and Beyond
Once strictly the realm of science fiction, conversational AI is increasingly making its way into people’s everyday lives. 2019 is already shaping up to be a milestone year in the technology’s development, and will no doubt end with conversational AI firmly entrenched in multiple industries, changing how companies do business and how the average person lives.