The Limited Times

Now you can see non-English news...

Rabbit R1, the mini computer that makes it easy to use generative AI - Insights

2024-01-18T16:35:56.163Z

Highlights: Rabbit R1, the mini computer that makes it easy to use generative AI - Insights. Presented at CES 2024, the device combines two artificial intelligence models to create a super digital assistant that could change the way we look at smartphones. RabbitOS operating system allows OpenAI's Large Language Model ChatGPT to respond to simple voice commands to perform complex actions for the user, effectively transforming him into a "super digital assistant" Rabbit R1 is not the first device to interact with AI as an alternative to the smartphone presented last year.


Presented at CES 2024, the device combines two artificial intelligence models to create a super digital assistant that could change the way we look at smartphones (ANSA)


by Alessio Jacona*

Our beloved smartphone, that is, the object that we always carry with us and with which we manage almost our entire existence, could soon appear less important to us: among the many things seen at CES in Las Vegas, between flying cars and transparent screens, there 'it was in fact also an apparently harmless object, yet potentially capable of calling into question what today is the total digital supremacy of the phones we have in our pockets.

The orange rabbit

It's called Rabbit R1 and it's a little orange box with a touch screen measuring less than 3", a 360° rotatable camera, two microphones and a speaker, 4G connection, the ability to make video calls, a ring to manage the information on the screen and an activation button.

A small object that looks like a toy, that costs like a toy (199 dollars) and looks quite anonymous, but which was actually created with the ambition of changing the way we use apps and digital services, making the most of maximize the incredible potential of generative artificial intelligence.

Yes, because the mini computer - or rather the pocket companion, as Rabbit founder Jesse Lyu defines it - brings with its RabbitOS operating system a technology that allows OpenAI's Large Language Model ChatGPT (based on GPT) to respond to simple voice commands to perform complex actions for the user, effectively transforming him into a "super digital assistant".

«Our mission is to create a computer so simple to use that you don't have to learn to use it - explained Lyu in Las Vegas - the best way to achieve this is to abandon the current approach based on operating systems that use apps, and move to an approach centered on the use of natural language.

This is why we have built a computer that we call a companion, and which is capable of speaking, listening and above all carrying out tasks."

After the LLM that talks, the LAM that does

Away with smartphone screens, away with a thousand apps, one for each service, each requiring the effort to learn how it works.

The Large Language Model on the Rabbit R1 is accompanied by a LAM, or Large Action Model, that is, a system capable of copying and replicating human actions: together LLM and LAM interact with us, understand our requests and act using all the necessary apps and services : for example, to book a lunch in a fish restaurant in a location close to home but by the lake, making sure they also have a vegan menu, choosing the trains to reach the place and purchasing the tickets.

All with a single request made verbally, by pressing the activation button.

Now: demos should always be taken with a pinch of salt, because they take place in a protected environment and every action has been tried and tested before.

That said, one must admit that watching Jessy Lyu interact with her R1's LAM is something reminiscent of the best science fiction.

This is also thanks to the system's reaction times, which according to the CEO of

Rabbit guarantees a response ten times faster than the others seen so far, and which actually responds in real time.

It should also be remembered that the Rabbit R1 is not the first device to interact with AI as an alternative to the smartphone presented last year: there is also Human's AI PIN, a wearable device with voice, touch and gesture interface which uses GPT-4 and which however (when it arrives on the market) will cost 699 dollars plus a monthly subscription of 24 dollars per month.

An AI that anyone can train

To allow the R1 to use the services you have subscribed to, simply connect your accounts to Rabbit OS via a specific configuration page.

The same one from which you can access another potentially revolutionary feature: the teach mode, which is a training space where you can "show" the R1's AI how to do something (in the demo Lyu teaches it to generate images with Midjourney ) and then this is automatically added to the available features.

As if that wasn't enough, you can also use the R1's camera to frame an object or image and ask the AI ​​for information about them.

Something to leave you speechless.

"We didn't build the R1 to replace your smartphones, it's a new generation of devices designed from the beginning around artificial intelligence," Jesse Lyu said during his presentation, and perhaps he really means it.

However, a question arises spontaneously: if the Rabbit R1 (and other similar devices that we may see appear in the near future) will really work as well as they promise, who will still want to spend over a thousand euros on a top of the range smartphone?

*Journalist, innovation expert and curator of the ANSA.it Artificial Intelligence Observatory

Reproduction reserved © Copyright ANSA

Source: ansa

All news articles on 2024-01-18

Trends 24h

Latest

© Communities 2019 - Privacy

The information on this site is from external sources that are not under our control.
The inclusion of any links does not necessarily imply a recommendation or endorse the views expressed within them.