Meta Hologram vs a live camera in product demos

A realistic presenter image can carry the conversation while leaving the product itself unseen. If you are planning a demonstration around Meta's Hologram announcement, reserve actual camera or screen footage for the details you want a customer to verify.

For Meta Ray-Ban Display, Meta describes Hologram as a digital face created after a short capture in the Meta AI app. During the call, voice drives the avatar and the model infers expressions. Meta announced US Early Access on WhatsApp for later this fall. That is the rollout described in its September announcement, not a claim that the feature is already available everywhere.

A convincing face is not a camera pointed at the product

UploadVR's David Heaney tried Hologram at Meta Connect. He reported initially mistaking a caller's synthetic face for a selfie-camera feed. In his account of the Ray-Ban Display version, setup also captures clothing and background, and the call generates the face from microphone audio. His hands-on report is the source of that observation; we have not tried the device.

A viewer may see a plausible presenter without seeing what that presenter is holding now. The avatar's smile also should not be treated as a direct measurement of the person's current facial expression.

Heaney describes a separate VR implementation with additional tracking. Do not transfer the voice-only explanation to every version called Hologram. This article concerns the Ray-Ban Display calling feature.

Match each claim to a view that can support it

Consider a creator presenting a new label design for a small bottle. These are different claims, even if one sentence introduces them all:

A generated mockup can help propose a look. Label it as a mockup, and use a photographed sample when the claim is about the printed object. Likewise, a still screenshot shows a state; it cannot establish how quickly a software action completed.

A proposed 35-second demonstration plan

Here is an original content plan for that bottle-label example, not a feature tutorial or a demonstration we recorded:

The plan can be assembled with ordinary captures. It does not assume WhatsApp can switch these feeds, export an avatar recording or integrate with an editor. Check the actual capture and sharing options before promising a particular presentation format. If an avatar segment is unavailable, an ordinary presenter shot works for the same introduction.

In Creatos, image and text nodes can hold the two actual sample photos with their captions while you plan the comparison. Keep the source type in each title: "printed sample A," "printed sample B," or "generated mockup." The source label tells the audience which image supports the decision.