AI isn’t just a new application — it’s a new user interface, said Isaac Lyman, senior software engineer and author of Your First Year in Code, in a May StackOverflow blog post.
Lyman dismissed sci-fi notions that large language models and generative AI are somehow intelligent, calling them “model printers” that are just “our own babble reflected back to us.” Instead, we should be looking at their potential to be a new kind of user interface, she said.
“As a substitute for human beings, they are no match,” wrote Lyman. “But as a replacement for, say, a command-line interface? They are a huge improvement.
It’s an intriguing idea that led us to ask: How will Generative AI change the frontend and UI? Will chatbots take over every website?
AI on the frontend
AI won’t necessarily lead to everything becoming a chatbot, but AI is already changing some frontend functions, said Lee Robinson, head of developer relations at frontend platform Vercel. First, Robinson is seeing more organizations implement natural language processing capabilities in site search.
“On the one hand, we’re seeing more interfaces that have natural language as a voice,” Robinson told The New Stack. “Instead of manually entering by clicking through a series of filters and doing a series of manual searches, [users are] able to talk to the app, to talk to the software, in natural language: “find me size 11 green shoes with the Nike logo.” I can just type it and it works. We’re definitely seeing more of that.
That’s not the only change Vercel is seeing thanks to artificial intelligence. The second is to build AI into the product in ways that aren’t necessarily apparent to the end user.
“We are also seeing an increase in companies using large language models. They’re using AI under the hood to create a better product experience, but you might not realize it at first,” said Robinson. “It’s behind the scenes. It’s part of creating a better product experience. this is a rapidly growing portion of companies infusing AI into their existing product offering.
AI and APIs
Lyman argued that the sweet spot for AI is not the output box, but the API. And its best use is not for intelligence, but as a user interface between humans and computers.
“AI may not truly understand us, but it can convey our intentions to an API with reasonable accuracy and describe the results in a way that is understandable,” he said.
Postman co-founder and CEO Abhivan Asthana also predicted that AI will change the graphical user interface, in a recent blog post.
“One area where I see great potential is simplifying complicated graphical user interfaces,” Asthana wrote. “For complex tasks, graphical user interfaces often become difficult to use and actions hide behind rows of buttons, menus, shortcuts and procedures. They take years of training for people to become proficient, and even then, most people struggle with them. Generative AI trained on domain understanding has the potential to streamline such experiences.
These AI bots will not just chat, but will be deeply integrated into existing workflows through which humans interact with computers, he added.
“For example, bots will start helping with a user interface and data intensive tasks, interacting via voice and of course interacting via a chat mechanism,” he wrote. In this scenario, APIs become the “hands and legs that feed the ‘thinking’ that AI is doing, according to Asthana. However, that could require some changes to how APIs are delivered, she cautioned.
“Until now, we’ve mainly designed APIs for applications used by humans, but designing APIs for machines will become an increasingly important area,” Asthana wrote. “If you are the leader of an organization, what does that mean to you? Well, if your organization has no APIs or has poorly designed APIs, you are invisible to these bots.”
It’s a shift Postman is already seeing, Ankit Sobti, co-founder and CTO of API Platform, told The New Stack.
“We are seeing more and more AI-powered consumers,” Sobti said. “We’re using it to discover information or find different points of connection between systems; and our opinion is that [with] APIs, you have to be aware as a detail designer, be aware of both human consumers and AI consumers, as you build them.
One tangible way APIs need to change to support AI is to support streaming responses, Robinson said.
“These models can take some time to think of an answer and generate the answer,” he said. “Many APIs need to support being able to stream responses, because it might take 10 seconds, 30 seconds, a minute to actually generate that image or to write the blog post. So the API needs to be able and enabled to do it.
This will also require pairing with modern platforms that can support streaming for the best user experience, Robinson added.
However, other ways the APIs need to scale are still being worked out, Sobti said.
“I definitely see a point where we’re on the spectrum of being aware that there’s a new group of consumers who are consuming APIs now: How are design patterns changing? How are development practices changing? And how the implementation practices change around it, that’s something that we believe we need to see and discover as an ecosystem,” she said.
#change #frontend #APIs
Image Source : thenewstack.io
