Big Tech is driving the new UX design of AI | The AI Beat

It was an astounding, whiplash-inducing week of AI product news from Big Tech. 

A new generative AI version of Alexa with a new, custom-built large language model (LLM) marking a “massive transformation of the assistant we love.” Microsoft’s AI companion Copilot baked right into the Windows operating system (OS), with a view across all applications. Google’s Bard tapping directly into Gmail, Docs, Maps and more. 

And just days after announcing its DALL-E 3 new-and-improved image generator with support for text and typography, OpenAI dropped a surprise announcement this morning that ChatGPT will now support both voice prompts and image uploads from users. 

There’s plenty to unpack from these announcements. But one area stands out to me: The emerging and constantly-improving user experience (UX) design of these AI tools, products and platforms — that presents AI tools to a user that is AI-aware and creates a friendly experience that allows users to play with the raw materials of an AI model and generate new output. Not just in chatbots, but in image generators, copilot-workflows and personal assistant devices.

It’s clear that Big Tech companies like Amazon, Google, Microsoft and OpenAI are driving this trend, which I chatted about a couple of weeks ago with Cassie Kozyrkov, who recently left a decade-long role as Google’s chief decision scientist to strike out on her own. 

AI designed for consumers 

I had already been thinking about how ChatGPT’s debut in November 2022 had jump-started a whole new way of thinking about UX design in AI. ChatGPT’s simple, user-friendly interface both obscured the complexity of the AI under the hood while making it clear that yes, this is an AI — with users invited to play with it to get what they want. 

Kozyrkov has pointed out that previously, AI design wasn’t about targeting users at all, but the builders of AI systems. In fact, she wrote in a blog post, “having a nontechnical user notice the AI component would be as embarrassing as drawing that user’s attention to issues of JavaScript versus HTML.” 

These days, in her speeches, Kozyrkov says that GenAI is actually “a UX revolution, not an AI revolution.” 

AI was traditionally a “subtle, unobtrusive component in software applications,” she explained. Now AI is being put in users’ hands to turn into anything they want. AI is the raw materials for creativity and productivity, she explained, and people can take it and use it in creative and interesting ways. 

Big Tech has the interdisciplinary teams

In an interview after the AI Native conference in New York City, Kozyrkov told me that too many people treat AI as a traditional research area where “one person does it all — that is such a myth.” That is not an easy space for UX people to operate in, she explained. 

But the truth is, AI is now a space where there are large, interdisciplinary teams, she continued. Big Tech companies, like Microsoft, Google, Amazon and OpenAI, are “lucky enough to be able to hire these interdisciplinary folk,” she said. 

Going forward, she predicted that things will change throughout the AI industry. “I hate to make predictions, but sometimes I’m confident enough to put one out there,” she said. “I think we’re going to go from engineering being the most effort to engineering being relatively less effort and design being more effort.” 

It’s still early days, she cautioned: “When you think about what it means to design these systems? What are we designing? Are we designing the tools and approaches so that a leader’s wishes can be implemented? Then there is how are we designing from the users perspective and representing users in different user populations and actually building the system in collaboration with design?”

A ‘radical pivot’ in AI-powered product design

There’s no doubt that what is happening today is a “radical pivot in product philosophy,” as users are encouraged to interact directly with AI components, Kozyrkov has said. And that is exactly what strikes me about the current Big Tech announcements. 

Generative AI is quickly becoming part of our workflow, our creative journeys, our daily lives, but with user experiences that are clearly AI-driven, not AI-hidden: Whether it is Amazon’s Alexa, Microsoft’s Copilot, Google’s Bard of OpenAI’s ChatGPT, users are AI-aware. And whether they are inputting text, images, or their own voices, they are judging the usefulness and impact of these products for themselves as they experiment — all while knowing that, while they can’t see under the hood, what they are experiencing is, indeed, the power of AI.