Study what people are saying about Rabbit R1 and take the following notes (just some observations, please don’t sue me 🙂. I believe every great company has an open mindset to receive some criticism).
Background on the Founder and Company:
The founder and the company previously worked on an NFT project two years ago, which integrated NFT, clean energy, brain control, and AI. The project failed.
At the time of the NFT project, the company was named Cyber Manufacture. It rebranded to its current name two months before officially announcing Rabbit R1.
More about the founder Jesse Lyu,if you interested
He graduated from a joint venture between the University of Liverpool and Xi'an Jiaotong University. Students from the uni typically spend 2 years in China and 2 years in the UK for a 4-year program.
He started Raven Tech, which was sold to Baidu in 2017
Criticisms on LAM:
Rabbit R1 operates on hard-coded scripts (pre-written), rather than AI.
The term “LAM” (Large Action Model) introduced by Rabbit R1 contradicts the fundamentals of modern network mechanisms:
LAM is essentially a bot that claims to interact with user interfaces. However, the concept of bots is not new.
Service providers have long combated bots to ensure real human interaction, using tools like CAPTCHA, IP analysis, and behavioral analysis.
The current solutions for these anti-bot measures aren't cutting it, and LAM doesn't seem to address these fundamental issues either.
Other things
A leak on GitHub suggests that Rabbit R1’s Large Action Model is not AI-driven but relies on Playwright automation scripts.
Xyzeva claims in a post that Rabbit R1 is breaking the Android license, censoring their research, and lying to the community. A very serious accusation.
The "LAM" code in Rabbit's Uber section includes a feature flag that enables someone in the backend to manually solve captchas while the system is running.