From CES Star to Financial Struggle: What Went Wrong with the Rabbit r1?
If one product were to be credited with catapulting "AI hardware" into the public consciousness, the Rabbit r1 would undoubtedly be a top contender. When it debuted at CES 2024, this small, orange device captured the world's attention with its retro design by Teenage Engineering, its large model-powered "natural language operating system" rabbitOS, and its bold promise to "replace apps and control all services with a single sentence."
However, this once-celebrated company, hailed as an industry pioneer, is now reportedly facing severe financial distress, with rumors of unpaid salaries for several months and strikes by some employees and outsourced teams. Despite the news of financial strain, Rabbit's official website remains operational, even offering the r1 at a discounted price of $159 with a 30-day return policy. This suggests the project hasn't been completely scrapped, unlike the Humane AI Pin. Rabbit's CEO, Jesse Lyu, has acknowledged the severe financial pressure, attributing it to setbacks in entering the Indian market—a challenging move considering the country's average monthly salary is around $250, making a $199 AI gadget a luxury for most.
Rabbit's current predicament has led many to adopt a pessimistic view of AI hardware's future. Yet, it's far too early to make such a judgment. The rise of large AI models is a recent phenomenon, and AI hardware is still in its infancy. The failure of one company does not signify the failure of an entire category.
Did Rabbit Fall Because It Took Too Big a Leap?
During the AI frenzy of 2024, a mere concept could attract immense investment and attention. Rabbit, with a tangible product ready for sale, became the focal point of the AI race, with many calling it the "next Apple" or heralding another "iPhone moment." Initially, Rabbit delivered promising results, announcing over 100,000 units of its first-generation product sold by August 2024, making it the best-selling AI hardware at the time. The excitement was fueled by near-magical product demos, but as the hype faded, users discovered a significant gap between promise and reality. The core issue wasn't just battery life; it was that many key features showcased in the demos were simply not available. The true reason for its failure was its attempt to build a "smartphone replacement" in an immature market. It overreached.
Rabbit's ambition to use its Large Action Model for complex tasks like booking flights often resulted in a clunky experience that took minutes, compared to seconds on a smartphone. The device provided more "emotional value" than real efficiency. Furthermore, its hardware design was flawed: the screen was too small for complex information, and the camera, a key feature, lacked a killer "AI vision" application. This combination of hardware raised costs, while the lack of a standout, indispensable feature left it in an awkward position. Its business model, which relied on costly commercial AI models without a subscription fee, was also unsustainable. The dream of an "Appless" ecosystem proved to be a fantasy, forcing Rabbit to later introduce "Creations," which allowed users to train automation scripts—an admission that it couldn't escape the existing app ecosystem. In essence, Rabbit's failure wasn't because AI hardware is a flawed concept, but because it tried to innovate interaction, rebuild the hardware industry, and reshape the software ecosystem all at once—a monumental task for any startup.
Chinese Manufacturers Have Their Own Ideas About AI Hardware
Does the cooling interest in general-purpose AI devices like Rabbit spell doom for the category? Not at all. The real question is how AI hardware should integrate into the existing smart ecosystem. Instead of aiming to be the "next smartphone," a better approach might be to perfect the experience for a specific group of users or a particular scenario. Chinese manufacturers are already exploring this path. For instance, the office software company DingTalk released the DingTalk A1, a device focused entirely on meeting recording and office collaboration. Essentially an ultra-thin voice recorder, it seamlessly integrates with the DingTalk ecosystem to automatically transcribe recordings and generate project tasks. Similarly, AI glasses and AI earphones are emerging, focusing on specific functions like photo translation and real-time audio enhancement, respectively. Compared to the ambitious leaps taken abroad, Chinese companies are adopting a more conservative strategy, focusing on augmenting existing hardware forms to cultivate user habits before moving toward more complex, multi-functional AI devices.
AI Hardware Shouldn't Aim to Replace Smartphones
A key characteristic of the approach in China is that AI hardware is designed to complement, not replace, smartphones. Ignoring the world's most ubiquitous smart device and its mature ecosystem is a flawed strategy. Leveraging a phone's app ecosystem and on-device computing power can significantly enhance the AI hardware experience. Rabbit's biggest mistake was its decision to go head-to-head with the smartphone, an established titan. It seems Rabbit may have learned this lesson, as it claims its next-generation product will be a more complete "three-in-one" device. For now, major breakthroughs in AI hardware are unlikely to happen overnight. The current trend is a gradual one, focusing on refining user experiences in specific scenarios. Perhaps in a few years, we'll find that everything around us has become "AI hardware," and the debate over hardware ecosystems will be a thing of the past because AI will be everywhere.
