On-device reactions
Oneira's expressive reactions run on the device, so its face can respond without relying on a cloud request.
See the local corePrivacy guide / intentional input
Compare always-listening AI wearables with Oneira's intentional approach: on-device expressions, owner-triggered Sight, and explicit connected requests.
Product status
Working today means the expressive face has been demonstrated on a real Oneira device.
Connected beta features require the Oneira app, internet access, a compatible service, and permissions you approve. They can change and may not work with every service.
The short answer

Oneira Dream Big+
Small enough to look playful. Built with a dual AI chip, OLED screen, and 1K camera, plus optional connected experiences you control.
Each $80 Founding Unit includes one Oneira device with a dual AI chip, OLED screen, and 1K camera, plus the selected color and included eye set. The 1K camera is included; its final production module and other non-material production details may change before fulfillment. Optional connected services, provider accounts, usage, and third-party fees are separate.
Oneira's expressive reactions run on the device, so its face can respond without relying on a cloud request.
See the local coreThe working prototype includes a dual ai chip, oled screen, and 1k camera. The 1K camera is included in the $80 Founding Unit; its final production module may change before fulfillment.
Inspect the hardwareOptional compatible connected services can extend Oneira when you choose. Setup, permissions, internet access, provider terms, and usage costs remain separate. Connected experiences work only with compatible services the owner deliberately configures. Oneira does not connect automatically to every service.
See compatible connectionsCompatible connected tools are optional and owner-configured. Nothing is preconfigured, and universal compatibility is not promised.
See connected featuresThe 1K camera supports deliberate, owner-triggered experiences. Optional Dream creation can transform a photo when the owner chooses it.
Open Oneira CreateOptional helper experiences have narrow, visible limits. Broader research, coding, and workspace actions remain in development.
Understand the agent pathA simple flow
Motion changes the expression without ambient capture
Deliberately start Sight or a connected request
See the destination, response, and important limits
The difference
A small companion can feel responsive because its face reacts to orientation and movement. It does not need to capture every conversation or continuously interpret the room to create that moment.
Oneira separates local expression from connected intelligence so the start of a camera or cloud interaction can be obvious.
Camera boundary
Oneira Sight processes a single owner-triggered frame in notice, read, or help mode, and the response is short enough to review on the device. Everyone present must agree before an owner-triggered Sight capture.
JML does not store the Sight frame or reply in this release. The connected request remains subject to the active service's policy.
Buyer checklist
Look for a hardware indicator, a clear activation gesture, a real off state, retention disclosures, bystander consent guidance, and a list of destinations that receive data.
A privacy promise is strongest when the product architecture makes unwanted capture unnecessary, not merely when a policy promises careful handling later.
Privacy-first companion / understand the idle state
“Not always listening” can mean different things across consumer devices, so evaluate the concrete idle behavior instead of relying on a soft privacy label. Oneira’s core character responds to movement on the device: it can react when carried, tilted, moved, or set down without turning that physical context into a continuous conversation. Its everyday experience is deliberately not presented as an always-open microphone, a passive camera feed, or a background surveillance system.
Visual help has its own visible boundary. Oneira Sight is Connected beta, owner-triggered, and built around one intentional frame rather than ongoing capture. Everyone present must agree before an owner-triggered Sight capture. JML does not store the Sight frame or reply in this release, while the active connected service's controls still apply. That distinction is useful even for people who never use Sight: it tells you which behavior lives on the hardware and when a connected request begins.
The practical test is whether you can explain the device’s quiet state to someone sitting nearby. A clear answer makes a companion easier to carry in shared spaces: movement can shape its on-device expression, while a connected request has an intentional beginning and visible consent requirement.
Decision framework
Questions, answered
No. Oneira is not marketed as an ambient listener. Everyday expressions use motion input, and connected requests require an intentional action.
No. Sight uses one owner-triggered frame and does not record video in the background.
The owner must deliberately start capture. Everyone present must agree before an owner-triggered Sight capture. Product and app cues should make the capture boundary clear.
No. Core motion reactions run on-device, but Sight and optional connected services require the app, internet access, and connected processing.
Oneira / current availability
500 Drop 01 units remain across 5 currently available colors. Availability is confirmed when secure checkout begins.
Preorder a Founding Unit · $80Oneira's focused helpers have narrow, visible limits. Broader research, coding, and approved-workspace actions remain in development.
Look for a plain explanation of what the product senses when you are doing nothing. Motion-aware reactions are different from continuous audio capture or camera recording, and a trustworthy device should not make buyers infer that distinction from marketing language.
A deliberate press, mode choice, or capture action makes the start of a connected request easier to recognize and control. If a device acts from ambient context, ask how you can tell what it collected and when the collection stopped.
Privacy is not only about the owner. Everyone present must agree before an owner-triggered Sight capture. Consider homes, offices, classrooms, and shared spaces where another person may not want to appear in a connected visual request.
Once a request is connected, identify which policy applies and what account controls are available. JML does not store the Sight frame or reply in this release, but the active connected service's rules still apply.
Good fit
Choose differently