We were first introduced to the Rabbit R1 in January 2024, at CES 2024, but what exactly is it? The charming sidekick (designed by Teenage Engineering) promises to take pocket gadgetry to the next level – offering something like a smartphone, but with an intuitive, unified, AI-driven interface that means (at least theoretically) interacting with individual apps and websites.
If you’re curious about the Rabbit R1 and the ways it could change the course of personal computing – or at least show us how the next generation of smartphones’ voice assistants could work – we’ve rounded up everything you need to know about it know here. From what it costs and how it works, to the AI engine that powers the R1 experience, all the details of this potentially revolutionary device can be found here.
The first batches of the Rabbit R1 will start shipping to users later in 2024, although it looks like availability will be quite limited at first – so you may have to wait a while to get your own Rabbit R1.
Rabbit R1: One minute overview
The Rabbit R1 is a lot like a phone in appearance, and in features: it has a camera and a SIM card slot, and it supports Wi-Fi and Bluetooth. What’s different, and what makes the Rabbit R1 special, is the interface: instead of a grid of apps, you get an AI assistant that talks to your favorite apps and does everything for you.
For example, you can let the R1 find a holiday destination and book flights there, queue up a playlist of your favorite music or book a taxi for you. In theory, you can do everything you can already do on your phone, just by asking. That said, many questions remain about how exactly it works and how it protects your privacy in the way it describes.
We’ve seen next-generation personal assistants in movies like Herand the R1 aims to make that a reality – using the latest AI capabilities to replace the traditional smartphone interface with something much more intuitive and slick.
Meet r1. Watch the keynote. Order now: https://t.co/R3sOtVWoJ5 #CES2024 pic.twitter.com/niUmjFvKvEJanuary 9, 2024
Another way to think about the Rabbit R1 is as an evolution of the Amazon Echo, Google Nest, and Apple HomePod smart speakers. The voice-activated digital assistants on these devices can perform some rudimentary tasks – like checking the weather or playing music – but the R1 aims to go far beyond what they’re capable of.
Rabbit says the R1 is “the future of human-machine interfaces,” and you can watch the pitch for the device in the heavily Apple-flavored CES 2024 keynote below.
Rabbit r1: release date and price
The first batch of 10,000 units of the Rabbit R1 was made available for pre-order at the same time the device was announced at CES, on January 9, 2024. Those units quickly sold out, as did a second batch of 10,000 units that became available shortly. after.
Rabbit says that the people who received their pre-orders should have their devices shipped to them in March and April 2024. At the time of writing, there’s no indication yet of when a new batch of units will be made available for pre-order, or when we might do so. watch the R1 go on sale on a larger scale.
What we do know is that the price of the Rabbit R1 starts at $199, which works out to around £155 / AU$300. To start, the Rabbit R1 is available to order in the US, Canada, UK, Denmark, France, Germany, Ireland, Italy, Netherlands, Spain, Sweden, South Korea and Japan, starting Rabbit’s website.
Plus, unlike competing AI devices like the Humane AI Pin, there are no ongoing subscription fees to pay.
Rabbit r1: hardware
The Rabbit r1 is square and bright orange and comes with a 2.88-inch color touchscreen on the front. It’s a fairly compact device, almost small enough to fit in the palm of a hand, and weighs just 115 grams (about 4 oz). There is only one design for now – you can’t choose this in multiple colors.
We know there is a far-field microphone embedded in the R1, as well as built-in speakers. There’s also an integrated 360-degree camera here, which is apparently called the Rabbit Eye. You can interact with elements by touching the screen, and there’s also an analog scroll wheel on the side of the device, if you need it.
On the right side of the Rabbit R1 is a push-to-talk button, which you use when you want to talk to the AI assistant on the device. There’s no need for a “hey Google” or “hey Siri” wake-up command, and it also means the assistant doesn’t have to constantly listen to your voice. Double-tapping the button activates the built-in camera.
Under the hood we have a 2.3GHz MediaTek Helio processor, and Rabbit says the device offers “all day” battery life. That battery can be charged with a USB-C charging cable and power adapter, but note that these aren’t included in the box, so you’ll have to use the ones you already have.
Rabbit R1: software
With its bright orange casing, the Rabbit r1 looks cute, but it’s the software that really makes it stand out. If you’ve already used something like ChatGPT or Google Bard, this is something similar: Rabbit OS is run by an AI chatbot, which can both answer questions and perform tasks.
In the CES keynote demo, Rabbit founder and CEO Jesse Lyu showed off the R1 while answering philosophical questions, checking stock prices, looking up information about movies, playing music on Spotify, booking an Uber, ordering a pizza and planning a vacation (completely with flights and hotel reservations).
To get some of this to work you’ll need to connect the Rabbit OS to your various apps and services, which can be done via a web portal. From the demo we saw, it looks like Spotify, Apple Music, YouTube Music, Expedia, Uber, eBay and Amazon are some of the services you can connect to.
Rabbit is keen to emphasize that it doesn’t store your login details or track you in any way – it simply connects you to the apps you need – although the details of how it does this via the cloud are still unclear.
Rabbit’s privacy page gives us a few more details, stating that “when you interact with Rabbit OS, you will be assigned a special and isolated environment in our cloud for your own LAM (large action model). When our rabbits perform tasks for you, they will use your own accounts that you have given us secure control over through our rabbit hole web portal”.
It also adds that “we do not store your passwords for these services. Rabbits will ask for permission and clarification while performing tasks, especially when it comes to sensitive actions such as payments.” How exactly Rabbit provides each user with a “dedicated and isolated environment” in its cloud isn’t clear yet, but we should find out more once it reaches the first early adopters.
We’re also told that the R1 can handle communications, real-time translation and analyzing images captured by the camera – for example, show the R1 what’s in your fridge, and it can come up with a dish for you to cook.
The Rabbit R1 also promises fast responses, faster than you’d get with other generative AI bots. You can talk to the R1 as you would Siri or Google Assistant, or you can summon an on-screen keyboard by shaking the device. It calls the built-in AI a Large Action Model or LAM, similar to a Large Language Model or LLM (known from bots as ChatGPT), but with much more freedom of choice.
Additionally, Rabbit says you can teach the R1 new skills. So if you show him how to order your groceries online, he can do it all by himself next time. In the CES demo we saw the Rabbit R1 learn how to create AI images via Midjourney, and then replicate the process itself.
Interestingly enough, Rabbit says it doesn’t want the R1 to replace your phone – it wants to work alongside it. For example, the R1 can’t surf YouTube, check social media, or organize your email (at least not yet), so it looks like the humble smartphone will be around for a while.
While some details about how the Rabbit R1 works and interacts with your favorite apps and services remain unclear, it’s undoubtedly one of the most exciting pieces of AI hardware yet – as evidenced by the rapid sell-out of early stock. We’ll bring you more first impressions as soon as we get our hands on one of 2024’s early tech stars.