Motorola’s new LAM is far more thrilling than the Rabbit R1 ever was

My irritating experiences with the pocket-sized, hype-fueled Rabbit R1 are well-known at this level. From letting it wrestle to information me via a protracted day to trying to train it new methods, it’s let me down every time I’ve given it an opportunity. And but, I’m nonetheless intrigued by the thought of an AI-powered mannequin that may study my intentions and do easy duties for me. I’d love to provide a voice command and have a Giant Motion Mannequin name me a journey to the prepare station or order my morning espresso.

So, when Motorola introduced a LAM of its personal at Lenovo Tech World, I used to be desperate to see it in motion. I wished to know if it felt extra upfront and reliable than the Leuchtorange-colored R1 gathering mud on my shelf. It solely took a couple of months for me to get my likelihood, and right here’s how Motorola received me over proper off the bat.

Right here’s the deal — I received to take a look at Motorola’s LAM in motion on the tail finish of CES 2025, so that you’ll need to put up with a playing reference or two. And now that we’re on the identical web page, let me stroll you thru the demo I received to expertise.

Motorola’s LAM expertise began as merely as anything on the Razr Plus (2024): I watched a Motorola worker open her flip cellphone and faucet on an app. This time, she wasn’t tapping on one thing like Uber or the Starbucks app; she was opening Moto AI. For now, that’s the place Motorola’s LAM lives — in its personal devoted app the place builders can slowly however steadily fine-tune its coaching.

As soon as there, she requested her Razr Plus to order her a “espresso Americano.” She didn’t ask for a particular measurement or specify the place she wished it from — only a drink to be prepared for pickup. And as quickly as she completed her command, the LAM sprang into motion. It jumped from the Moto AI interface to the Starbucks app and set about inserting her order for pickup on the nearest location (The Venetian foyer just some hundred toes away). She, in fact, canceled the order earlier than it might undergo, as no one wants a venti Americano each time they do a LAM demo, not even at CES.

Her level was made: Motorola’s LAM works shortly and retains its playing cards on the desk. Somewhat than logging into companies like Doordash and Spotify by way of a sketchy backdoor digital machine within the Rabbithole, you merely obtain apps to your cellphone and perform a little educating alongside the way in which. When you’re consistently choosing Uber XLs or selecting Consolation Electrical rides, Motorola’s mannequin will study that after a couple of tries and emulate your preferences.

Extra importantly, Motorola’s mannequin didn’t really feel prefer it was playing. I watched the Motorola worker frantically attempt to cancel an Uber journey to the Bellagio just because it picked out her journey and located her a driver sooner than I’ve ever seen a journey secured in Las Vegas. It didn’t brush her off with an excuse about digital rabbits being too drained or power her to make use of a little bit scroll wheel to make her selections; the LAM simply paid out its winnings as a result of it knew what she wished.

As you may in all probability inform, Motorola’s LAM proved extra spectacular to me at each flip. Do I like that it nonetheless takes over your complete show of the Razr Plus (2024) to “carry out actions,” as Motorola places it? No, not likely — I’d like to see it graduate to run within the background so I can proceed to make use of my cellphone within the meantime, but it surely’s a begin. In addition to, by watching the mannequin open Uber and hail a journey, I no less than know that it’s doing what I need and the way I need it executed.

Maybe what provides me confidence within the usefulness of Motorola’s LAM is that, nicely, it’s form of an app. Rabbit made an enormous fuss about how its Rabbit OS is decidedly not an app and might, subsequently, be extra helpful than an app — no less than till it was confirmed to be an app. And with Rabbit’s not-an-app-app, the restrictions had been instantly obvious. It might solely work with a brief checklist of digital machine-based integrations (Doordash, Spotify, and others). In any other case, it could run via a variation on an OpenAI mannequin when requested to seek for a solution.

With Motorola’s mannequin, nonetheless, it nonetheless feels just like the door is vast open. Sure, the examples of Uber and Starbucks I used to be proven had been quite simple and particular, but it surely felt just like the LAM was in management from begin to end. I didn’t need to examine in and information it alongside the way in which; it merely discovered the process over time and replicated it — no difficult Train Mode or accident-prone LAM Playground required. And, as a result of it makes use of the apps already in your cellphone, it feels more likely that Motorola’s mannequin will develop and develop to incorporate ordering from Amazon, mapping out eating places in Google Maps, and different issues that I may be too lazy to do.

After all, there are nonetheless a couple of issues that I’ll have to attend on with regards to Motorola’s AI-powered LAM. For starters, I’m but to learn the way the coaching course of truly works. Though I noticed the mannequin run via some fundamental duties, the Motorola crew was a bit tighter-lipped on how the LAM (which is at the moment nonetheless in beta testing) would go about increasing its skillset. And but, I’m keen to take that threat if it implies that my main machine — not a secondary companion — can truly function the AI companion that Rabbit thought it was promoting me.