In the past couple of years we have already seen the sporadic implementation of conversational user interfaces. However, with recent advancements in natural language processing, consumer hardware and software, and the deep integration of conversational UIs in ecosystems we can expect that these types of interfaces will start to become more capable and ubiquitous in the coming year.
- The advent of wireless earbuds could help make virtual personal assistants ubiquitous. A few manufacturers have launched or are planning to launch wireless earbuds (e.g. Apple, Beats, Bragi, Samsung, Jabra, and Intel).
- Virtual Personal Assistants (VPAs) are increasingly integrated in products such as smartwatches, smart-speakers, desktops/laptops, set-top boxes and IoT/smarthome applications.
- Chatbots are revived with Microsoft and Facebook integrating conversational AIs in messaging applications as a way to plug into third party services intuitively.
- Gartner predicts that in 2019 20 percent of all user interactions will take place via VPAs.
All of the big tech companies are currently involved in conversational UIs, although with some notable differences. Apple seems to have a good position in terms of consumer hardware touch points (smartphone, smartwatch, earbuds, set-top box and computer), but its VPA Siri is less sophisticated when compared to the competition (e.g. Alexa, or Google Assistant). In part this seems to be caused by Apple’s constrained developer ecosystem, and stricter privacy policies which hampers the data they can collect to train their algorithms. In contrast, Amazon radically opens up their conversational tech through their AWS infrastructure (Amazon Lex) and third party hardware integration by providing a dev kit for home applications, which could accelerate the improvement of their conversational UI considerably.
In the end, the adoption of conversational UIs is part of the move towards more natural user interfaces in which we increase the bandwidth but also the intimacy with which we interact with computers (touch, gestures, gaze and even thought)
From a societal perspective, exposing our conversations ubiquitously to computers, especially through voice, could have significant consequences. One recent example is the hot tub murder in which the police are demanding access to the audio recordings of an Amazon Echo which was present at the scene. Here we should also realize that our voice can disclose much more about ourselves than we would like, ranging from telling lies, detecting mood and even uncovering incipient heart disease.
In the end, the adoption of conversational UIs is part of the move towards more natural user interfaces in which we increase the bandwidth but also the intimacy with which we interact with computers (touch, gestures, gaze and even thought). Furthermore, changes in the way people interact with computers in turn influence the way institutions, companies, services, and products relate to people. Moreover, these changes also influence the way people interact with each other as exchanges are becoming increasingly computer-mediated.