Artificial Intelligence | News | Insights | AiThority
[bsfp-cryptocurrency style=”widget-18″ align=”marquee” columns=”6″ coins=”selected” coins-count=”6″ coins-selected=”BTC,ETH,XRP,LTC,EOS,ADA,XLM,NEO,LTC,EOS,XEM,DASH,USDT,BNB,QTUM,XVG,ONT,ZEC,STEEM” currency=”USD” title=”Cryptocurrency Widget” show_title=”0″ icon=”” scheme=”light” bs-show-desktop=”1″ bs-show-tablet=”1″ bs-show-phone=”1″ custom-css-class=”” custom-id=”” css=”.vc_custom_1523079266073{margin-bottom: 0px !important;padding-top: 0px !important;padding-bottom: 0px !important;}”]

Thanks to Generative AI, Conversational UI is Poised to Reshape e-Commerce

By Matthew Gertner, Founder and CEO of Salsita

Now that the initial amazement over ChatGPT has subsided, it is becoming increasingly fashionable to express skepticism about its usefulness. “What are the new wave of LLMs actually good for?” wonder punters and pundits, referring to the large language models that power ChatGPT, Gemini, Claude, and others.

They’ve proved their worth many times over with time-saving functions such as summarizing text, writing boilerplate, and even generating a serviceable recipe for a mango margarita. But this technology was meant to change the world; surely there must be more to it.

ChatGPT and its ilk are poised to fundamentally change the way that people use computers by introducing a new user interface paradigm: Conversational UI or CUI. CUI replaces pointing, clicking, pinching, and tapping with natural conversations, thus enabling users to take advantage of our brains’ incredible language capabilities. Unless you were raised by wolves, you can also leverage the many years of experience you have communicating your thoughts and desires to other people–and understanding their responses–using natural language.

Also Read: Leveraging AI to Plug the Expanding $2.5 Trillion Trade Finance Gap

Know where we came from to understand how we got here

To fully grasp this trend, it is worth looking at previous paradigm shifts in user interface technology in order to understand how, and why, they occurred.

When personal computers burst onto the scene in the mid-1970s, they looked very different from what we are familiar with today. For example, the MITS Altair 680 – one of the first PC systems – does not have a particularly ergonomic user interface; data had to be inputted as binary code using toggle switches (up for 1, down for 0)with the computer responding using a row of LED lights.

Tech has come a long way since then. Toggle switches gave way to MS-DOS’s command line, which was supplanted a few years later by GUIs on Windows and the Macintosh. More recently, the iPhone popularized a high-tech touchscreen that introduced natural-feeling gestures like swiping to scroll and pinching to zoom.

To some degree, these improvements have been driven by an ever-evolving understanding of how best to interact and communicate with computers, something more commonly referred to nowadays as user experience (UX). But mostly, improvements in UX were gated by what was possible at the time. Though it’s unlikely that anyone thought that toggle switches and LEDs were the ultimate in ergonomics, they didn’t have the tech back then to do it better.

 Dirty tech truths: Expectations vs. reality

Related Posts
1 of 7,375

Oh, how times have changed. Recent tech advancements have brought the advent of Conversational UI. The term has been around for some time, predating the appearance of ChatGPT and other LLMs. A quick Google search turns up references from 2017, and the Wikipedia page for “Conversational user interface” dates back to 2019. Until now, however, this approach has failed to catch on for one simple reason: chat technology was still found to be lacking.

Though Hollywood and science fiction novelists alike have long theorized about how helpful and appealing it would be to have a computer able to communicate conversationally, the reality of pre-LLM technology was quite different. Users needed to jump through hoops to get the machine to understand even basic commands. More often than not, they were led down a maze of pre-defined options rather than holding a freeform conversation. When CUI resembles the ordeal of getting through your bank’s automated system to a live representative, it is not surprising that users aren’t exactly thrilled with the prospect.

Modern CUI is different. The current buzz and excitement around AI is a direct result of the amazing new capabilities demonstrated by ChatGPT when it burst onto the scene in late 2022. These capabilities make it possible to create a far superior conversational experience where the computer truly understands the nuance in what users are trying to say, and responds in clear, correct native language.

Also Read: Three Ways Generative AI Can Accelerate Knowledge Transfer Across An Organization

Not all CUI are created equal

It is important not to confuse CUI with other uses of LLMs. After ChatGPT was released, companies scrambled to add features to their existing software to do things like summarize text and generate new content. To say that this functionality was low-hanging fruit is an understatement: with a modern LLM, this was so straightforward to implement that the fruit was basically lying on the ground waiting to be picked up and eaten. The trouble was that it wasn’t proper CUI, since it doesn’t fundamentally change how users interact with the software.

Nor is CUI an all-or-nothing proposition. Some things will always be easier to do with a conventional UI; for example, it is typically easier to position something on the screen using a mouse or trackpad, rather than trying to explain to the computer how to do it. When absorbing large amounts of data, it is far better to let our visual cortex do its thing by using graphs, diagrams, and videos.

Back to the future

 Striking the right balance between conventional and conversational UI is still an art, not a science. Much has been made of the potential to give instructions like “book me a trip to NYC for a week this summer” and have the LLM do everything: choose the best dates, find the perfect hotel, get you tickets for the hot new Broadway show, book a fabulous dinner at the best omakase restaurant in town, and even order your spouse an expensive gift because they really wanted to spend the week on the beach until you overruled them. Maybe one day but, with current technology, it is more realistic to guide the user through the process interactively.

So, where does that leave us? Are we on the brink of a conversational utopia, or is this just another fleeting tech fad? Imagine trying to convince someone in the 70s that one day they’d be scrolling through cat memes on a touchscreen device while pretending to listen to a video conference on their 15″ color laptop. The truth is, we’re right at the start of an exciting transition. CUI isn’t just a novelty; it’s the latest step in a long journey towards more intuitive and natural human-computer interactions, powered by the jaw-dropping abilities of modern LLMs.

[To share your insights with us as part of editorial or sponsored content, please write to psen@itechseries.com]

 

Comments are closed.