Two years ago, AI wearables were going to replace smartphones. Then the Humane AI Pin launched, underwhelmed, and was eventually sold off. Rabbit's R1 shipped to disappointed pre-order customers who mostly returned them. Both devices shared a feature nobody actually asked for: they were always listening.
Two former Apple Vision Pro engineers are betting the problem wasn't the concept - it was the interaction model. Their device clips to clothing, looks like an iPod Shuffle, and does nothing until you physically tap it. No passive microphone. No ambient listening. Press it, it captures audio, you get a response.
It's a surprisingly restrained design choice for a product category defined by overreach. But restrained might be exactly right here. The AI Pin's failure wasn't just bad hardware - it was that people didn't want to walk around wearing a microphone at all times. That concern wasn't hypothetical; it was a purchase decision millions of potential buyers made by not buying.
The tap-to-activate model gives users a clear mental model: this device only knows what I decide to tell it. Neither the AI Pin nor the R1 could credibly make that claim.
The Vision Pro background is relevant in a specific way. That product also had to solve the question of when the device pays attention - whether eye tracking was actively processing what you looked at, when cameras were capturing, when spatial audio was recording. These are genuinely hard UX problems that most consumer hardware teams never face. A team that spent years working through those tradeoffs is better equipped to design a wearable that people will actually trust.
Pricing and release timing haven't been announced publicly. The product appears to be in early development.
The bar for success here isn't high. Every failed AI wearable had a compelling pitch. The real test is whether the device does something useful enough that people wear it every day - the privacy story gets you consideration, but utility is what closes the sale.