Rabbit, a tech startup, has introduced its latest hardware offering—an AI-driven walkie-talkie style device called the Rabbit r1. The device, which uses “AI agents” to perform tasks on behalf of the user, was showcased during a pre-recorded keynote address at the Consumer Electronics Show. The Rabbit r1 can plan vacations, order food, and book transportation services, all through natural language interaction. Priced at $199, the r1 competes with other portable AI-first devices like Humane Ai Pin and Meta and Rayban smart glasses.
The growing interest in AI-first hardware stems from the belief that advancements in AI, such as large language models, will revolutionize personalized computing. Tech investors, including OpenAI CEO Sam Altman and Softbank’s Masayoshi Son, are investing heavily in this field. Rabbit has raised $30 million in funding, led by venture capital firm Khosla Ventures.
What sets Rabbit’s r1 apart is its use of a new AI system called a “large action model,” which enables it to perform actions in the real world. Unlike traditional AI models, the large action model is trained on graphical user interfaces, allowing it to navigate human-designed interfaces and execute tasks effectively.
To use the Rabbit r1, users need to sign into their various accounts through Rabbit’s web portal. Rabbit’s AI agents, hosted on external servers, leverage these accounts to carry out requested actions. The company assures users that their data is secure and that user passwords are not stored. However, Rabbit does not disclose its industry partners for natural language intelligence.
Despite the emergence of AI-driven hardware, some experts question whether such devices are necessary when intelligent phones can potentially integrate similar capabilities. Francisco Jeronimo of IDC notes that although the ideas behind these products are promising, consumers may not require additional devices.
There are also concerns about the risks associated with empowering AI to perform real-world actions. Powerful agential AI systems pose unique risks, including the pursuit of dangerous objectives. A paper published by the Center for AI Safety warns of the vulnerabilities that would arise in a society heavily reliant on interacting AI agents.
Sam Altman from OpenAI acknowledges the importance of addressing safety concerns and emphasizes the need for cautious progress in enabling AI systems to take actions in the real world. He believes that thorough consideration by society is necessary to avoid potential harm.
In conclusion, Rabbit’s introduction of the r1 device intensifies the competition in the AI hardware market. As the industry moves towards developing AI systems capable of taking actions in the physical world, it becomes imperative to address potential risks and ensure responsible deployment of these technologies.