Considering the fact that he made his wealth by co-founding a company that introduced a new way to transfer money, or that he builds what many consider to be the car of the future, or that he’s busied himself of late with building a spacecraft to take human beings to the moon, it may seem strange that Elon Musk has become something of a de-facto spokesman for people who fear the future.
The Tesla and SpaceX founder is especially unnerved by what he considers to be a credible threat posed by artificial intelligence to disrupt our lives. And this isn’t just the standard fear that you often hear expressed about how robots will take all of our jobs, but rather a seemingly genuine fear that AI could destroy life as we know it.
In a Vanity Fair profile from a couple of years ago, it was explained that Musk’s interest in space exploration is partly due to his fear of artificial intelligence. In fact, Musk has said that “one reason we needed to colonize Mars [is] so that we’ll have a bolt-hole if A.I. goes rogue and turns on humanity.”
Now, it’s important to remember when you’re incorporating AI-powered tools into your UX design—specifically, when building a chatbot function—that one doesn’t have to share Musk’s level of paranoia about artificial intelligence going rogue to still harbor some unease about the robot on the other side of the screen.
“Although AI innovation comes in areas with measurable higher accuracy than humans, users still may have certain objections [to] using a smart application,” writes Slavo Glinsky, a digital product designer, and co-founder of UI Motion Kit.
With that in mind, Glinsky advises you let people maintain control of the product—drawing the analogy of how a self-driving car still has a steering wheel and an emergency brake. This will “allow people to interact with it as [with] other items” on the product’s screen.
Of course, part of putting people at ease with using newish technology—or least technology they don’t really understand—is to create an intuitive user experience with your chatbot. Here are some best practices you can follow to make sure that happens.
First off, it is likely that companies will be embracing AI and conversational user interfaces (CUI)—i.e. chatbots—in order to reduce the costs associated with employing humans, which gives some merit to all the fears about robots stealing our jobs. But it could also be utilized as a way of complementing the work of their human employees, taking some of the menial work off their hands, and letting the humans focus on the bigger issues.
“Integrating a customer support chatbot that can respond to common customer queries and provide assistance 24/7 significantly reduces employment costs for companies who implement the systems, as their customer support agents then only need to deal with the complex questions a chatbot is unable to handle,” notes Merve Postalcıoğlu, a senior UX and product designer.
But it’s important that your chatbot has a clear purpose in your product design—and that this purpose is especially clear to your users, Postalcıoğlu adds. She writes:
Chatbot designers should begin by identifying the value a chatbot will bring to the end user, and reference it throughout the design process. It's here that UX designers add great value in framing the scope of the project through user-centered design techniques, such as research and ideation.
Postalcıoğlu explains how when she designed a chatbot for a bank, she utilized contextual inquiry, which “is a semi-structured interview method to obtain information about the context of use, where users are first asked a set of standard questions and then observed and questioned while they work in their own environments,” according to the Usability Body of Knowledge website. For Postalcıoğlu, the contextual inquiry process “was an insightful way to understand real conversations between agents and customers, and it helped to define the purpose of our chatbot.”
In terms of why a chatbot would be incorporated into product design, Raluca Budiu, director of research for the Nielsen Norman Group, points to two primary reasons: either for customer service or for interaction, the latter of which is often available on a platform like Facebook Messenger.
“While customer-service bots are often text only, interaction bots combine text with visual UI elements as a method of interaction,” Budiu explains.
But while you may have a clearly defined purpose for your bot, you must determine its value proposition to your product design. Does it actually improve the user experience or is it just a cool addition that users will rarely use?
Budiu notes that with customer-service bots, “if some of the users’ questions can be successfully addressed in an automated manner, [then] the business will benefit.” However, “if the bot is too rudimentary, people will lose trust in the company and will feel ignored and unappreciated.”
Things are less clear-cut with interaction bots, according to Budiu.
“These bots simply replicate functionality that is already available on the web or in mobile apps,” she writes. “Is it worth spending time and money for this new channel? Unlikely—at least in the US and other countries where the traditional channels are already well-established.”
Budiu adds that while interaction bots can be useful for “power users” who repeat the same tasks often, “a well-designed website will have shortcuts in place to help power users” anyway. One example of this is an e-commerce site that keeps users logged in to the site to make repeated purchases easier.
With that in mind, Budiu suggests that you don’t get “overly ambitious” with your bots, but instead use them for simple tasks.
“Complexity is not well handled in the limited bot interface,” she notes.
However, once you have a defined purpose for your chatbot and have determined that it can bring added value to the end user, the key is to have a chatbot with an intuitive, simple design that ensures the best possible user experience. Here are several tips for improving the UX design for your bot, according to various industry experts.
Be clear to users that they are communicating with a bot and not a real person to help prevent frustration or confusion over a conversation that seems almost human, but not quite. By emphasizing this distinction, it “makes people a bit more forgiving and also helps in guiding you to drive a less complex conversation without elaborated language and difficult grammatical structures,” according to a blog post from Ergomania UX. This gives the chatbot “a chance to understand what we are trying to tell it.”
We’ve all sent enough texts to know that typing words out on a device can be fraught with mistakes—and not all of them can be blamed on that damn iPhone autocorrect. So, design your chatbot in a way that tolerates typos from its flawed human users that allows it to at least get the gist of what’s being asked of it.
A chatbot may not have human feelings, but that doesn’t mean it has to be void of personality. Amazon’s Alexa does not sound like a real person, but you can interact with her in a way that she can feel a little bit like a roommate, albeit one that doesn’t let dishes pile up in the kitchen sink. “Your chatbot is voice and representation of your business, carefully crafting a personality can help determine the overall experience it will provide to the customers,” notes the team at IdeaTheorem. This means setting the vocabulary and the tone of the voice that matches your brand—whether its formal, casual, or something in between the two.
One of the problems with going all out on a first date and wowing your partner with how amazing and romantic and creative you were in planning the date, is that you’ve set almost unrealistic expectations for the second date. Remember this when designing your chatbot, so that you set realistic expectations for what it can and can’t do – and be consistent. Don’t allow your chatbot to know how to respond to one question but fail to grasp another question that is very similar and that the user thinks it should be able to answer. “Rather than attempting to give an illusion of high-level NLP (natural language processing) and then failing, you should set correct expectations that match your chatbot’s abilities,” according to Eunji Seo at Chatbots Magazine.
While it can be frustrating for a user when a chatbot can’t answer a question, you can lessen the frustration through your UX design. At the moment, “NLP and AI capabilities are currently nowhere up to meeting user expectations, so buttons and quick replies are almost integral for keeping users on the happy path and away from too many apologetic error messages,” Seo explains. She refers to these buttons and quick replies as “guardrails” that can help “prevent users from encountering the dreaded ‘blinking cursor’ scenario, where users are faced with an empty chat box which [gives] no indication of where to go.”
These are just several (of the many) tips to consider when building your chatbot in order to improve the user experience. Although AI technology has come a long way in improving the UX of chatbots, it’s important not to lose the human element because the reality is that there are going to be times when your chatbot will not be able to help the user.
“Be honest about not understanding,” Nielsen Norman Group’s Raluca Budiu writes. “Offer an escape hatch in the form of a real human, a phone number, or a link to a different interaction channel.”