Why chatbots fail

In the tech industry, we're so eager for the next big thing to arrive that we jump very quickly to announce its arrival. Driven partly by genuine enthusiasm, partly by inherent capitalist incentives, we generate wave after wave of hype. And there's a special flavor to this zeal reserved for technologies that achieve that pinnacle of science fiction fantasy: human-like artificial intelligence.

One problem with the term “artificial intelligence” is that it’s used so carelessly. The current AI chatbot narrative revolves around the use of natural language processing (NLP), for example, but Google Search uses NLP in its search results. a long time ago .AI is a marketing term used to generate hype, and the technology media this believing in that .

This doesn't mean that a wide range of "AI technologies" won't create value for the user. But, as product designers, we must be able to distinguish that value from sales pitches. It wasn't long ago that Microsoft's Cortana, Amazon's Alexa, and Google Assistant were growing to change the world. The way people spoke back then. , It seemed inevitable that these digital assistants would become co-pilots in our lives, relieving us of their boredom.

The big technology companies They claimed victory. , But five years later, the trend is clear: while useful for certain things, robots haven't changed society in a significant way. This low adoption isn't the result of resistance from Luddite activists, restrictive government regulation, or fear of evil machines. The issue is one of inertia – there simply aren't many reasons to go out of your way to use the thing, as it turns out.

Changing human behavior is no trivial task. There is both a academic discipline as a self-help industry dedicated to understanding its nuances. Sometimes, technology companies strike gold and create products attractive enough to enter our daily lives – personal computers, smartphones, and social media – but, generally speaking, it takes a lot for people to change their habits. For a new technology to catch on, it needs to hit a specific note at the right pitch, at the right time. Billions of dollars have been spent trying to figure out how to consistently hit that note, but it remains elusive.

The industry has had some time to test and iterate on chatbots, and now there are some excellent guidelines design That's all around us. Most of these guidelines acknowledge the limitations of chatbots, as well as their strengths. There are some important UX issues with conversational interfaces that make it difficult... the statement that chatbots like ChatGPT or Bard They are the next big thing in technology.

Text only

With a chat interface, the text is presented linearly. Imagine holding a pencil in your hand and trying to discern its color. If you're looking at the tip of the pencil, it will be difficult. If you can move the pencil and look at it from various angles, it will be much easier.

Using a chatbot is like looking at the nose of a pencil. When you ask a question and get an answer in text form, you get one piece of information at a time. Instead of typing a query into a search box and getting a list of answers in the form of text and other user interface elements like buttons and tabs, you get a text response that may or may not be useful.

That's why I'm very skeptical about it. New Bing search engine using ChatGPT technology. .When we're searching for something on the internet, there isn't necessarily one right answer. Sometimes ChatGPT returns a list of results, which helps a little. But a simple sorted list still lacks interactivity.

Over the years, even Google Search has evolved from lists to a much more robust user interface with tabs, expandable cards, and summary elements known as knowledge panels .These interactive elements put more control in the user's hands, allowing them to find the answer among the results, rather than relying on a chatbot to nail it with a conversational response.

Ambiguous entry

Another problem with chatbots is that you never know exactly what to say to get the answer you're looking for. That's why, if you've ever used a digital assistant, it first offers suggestions on what it can do. Without these prompts, users have no idea how to interact with it.

It's true that ChatGPT helps solve this problem because it has an incredibly sophisticated model that can respond very naturally to most user-generated messages. But the fact remains: if you're looking for a nuanced answer to a complex question, it's frustrating to go back and forth with a chatbot until you find what you're looking for. It's much better to get all the information in an instant and quickly find what's most relevant yourself.

Graphical user interfaces They were actually invented to solve that exact problem. In the early days of computing, you needed... command-line interfaces to perform functions. When the people at Xerox PARC invented the GUI, they empowered an entire population of non-programmers to use computers, recreating familiar models (e.g., files, folders) on a screen.

Novelty is not enough.

The main value of a chatbot, or any AI, is its novelty factor. People love to play around with it – it's fun! But if freshness is the main The reason people are using it is that it's a problem for the product's longevity.

Calling something “artificially intelligent” is arrogant. And the public’s response to that will be to immediately test the limits of this supposedly intelligent thing, hoping to break it. (Do I need to remind you of the infamous...?) bot Tay ?)

We've already seen journalists generate cycles of negative news for ChatGPT. doing that .In addition to the harmful rhetoric that unrestricted chatbots could spread, it's probably not good for business if the primary way people engage with your product is by trying to prove that it doesn't work very well.

ChatGPT is being touted as groundbreaking technology, but chatbots are actually an outdated idea. Our attachment to the fantasy of sentient AI is what makes this big news, while user experience is obscured and the hype train keeps rolling. We need to break this mold to unlock technology that will truly add value for users. What excites me personally isn't artificial intelligence (whatever that means), but what it can enable. This requires a trial-and-error process based on user needs and behavior, not hype. Strong human-centered thinking can steer technology in the right direction – and that's where the design community comes in.

If you're someone who enjoys reading a lot of semi-useful information in text form, I simply suggest... Subscribe to my Substack , where I write about a variety of technology and design topics. I promise you'll never be disappointed. manipulated in a sinister way to end their marriage. .😊