A new era of natural experiences
In the last few weeks, we've seen visions for the future of user and customer experience from the two most prominent AI players: Open AI and Google.
We're entering a new era of natural experiences.
This week, we'll examine the four characteristics of natural experiences: conversational, multimodal, agentive, and embedded.
We'll also examine the broad history of user experiences to understand where we're going.
From mechanical to natural user experiences
The first version of a mechanical computer was the punch card tabulating machine.
In the 1880s, a young engineer named Herman Hollerith invented it to solve the most significant computing challenge at the time —the US census.
In 1880, manually documenting the inhabitants of the rapidly growing young nation took seven years. Automating that work with punch cards made it dramatically more efficient, reducing the time to two years in 1890.
The history of human-computer interaction has been primarily mechanical. From the punch card to command-line interfaces and later computers with graphical UIs, we've made slow progress towards more natural experiences.
Now, we're entering an era of natural experiences.
Three underlying technologies drive it:
Large multimodal generative AI models like GPT-4o can reason across text, audio, and video
Portable super-computers known as smartphones
Augmented Reality devices that overlay the gen AI models onto the real world
These technologies enable dramatically different, more natural interactions with computers.
The four emerging principles of natural user experiences are:
Conversational - we get stuff done by the most natural user interface - natural language.
Multimodal: We can interact with audio, text, images, and video as we see fit. AI systems can" see" and" hear" the world.
Agentive - AI systems take care of actions requiring planning and several steps to complete
Embedded - the AI assistants can be accessed from smartwatches to cars and speakers.
Conversational
Increasingly, we interact with computers through the most natural user interface—conversation. This often simplifies complex issues better than a graphical UI.
Instead of looking at data dashboards about our fitness, we'll ask how to improve.
Natural audio conversations with almost zero latency will further drive this trend.
Healthify is a multimodal and conversational nutritional coach powered by generative AI
2. Multimodal
Computers will increasingly understand and produce text, audio, images, and video like humans.
They can also access their reasoning powers across these modalities.
Here are some glimpses into this future:
An AI assistant listens in on a meeting and, when asked, can summarize it or suggest the next action points. Open AI has already showcased this in one of their demos.
An engineer fixing a complex machine can point their phone video at it to receive instructions in real-time.
OpenAI launched GPT-4o with a vision for interacting with their AI-agent in a natural discussion and the camera of your phone
3. Agentive
New AI assistants will require less micromanaging.
Instead, we'll give them a goal - like plan next week offsite- and they'll break the goal down into sub-tasks and execute.
This agency is the explicitly stated vision of Google and Open AI. Google already showed specific examples of this rolling out in search at Google IO last week.
Google leaned heavily on agentive AI experiences in their Google IO a few weeks ago.
4. Embedded
This natural user experience will be embedded everywhere with smaller and cheaper chips.
We can access them from our watches, glasses, and cars. Natural conversations with AI agents will flow from one device to the next.
AI-enabled natural experiences will be embedded everywhere from personal wearable devices to our environment.
Updating our intuition
For the last 15 years, designing digital experiences has mainly been the same.
We've built digital services programmed to do specific things via a visual UI.
Now, we'll have to start updating the OS of our intuitions.
We're now designing natural experiences - conversational, multimodal, agentive, and embedded.
We'll have to experiment with designing using this new lens. The next 5 to 10 years will be a period of experimentation and discontinuity. New winners and losers will emerge.
Like with punch cards and smartphones, the ones who embrace this new world with measured quickness will thrive.
——-
PS. A quick announcement:
I'm excited to launch Designer 2.0 Live - a new Gen AI training workshop for UX, product, and service designers. The workshop will give designers principles and ideas for amplifying their creativity and efficiency with AI.
I've run a tailored version of this training for 600+ designers from companies like KONE and Futurice. Now, anyone can join the training remotely on Oct 24, 10:00 - 16:00 GMT+3.
Early bird tickets are available now for 690€+VAT until June 30.
Hope to see you there!
