Live, recorded or AI?
Three ways to fill a hologram display. The right one depends on who must be there and who will ask what.
Use live when a particular person must speak and respond now. Use recorded when the message is fixed and should play on repeat, with no one to run it. Use an AI avatar when visitors will ask many different questions over a long period.
Side by side
| Live | Recorded | AI avatar | |
|---|---|---|---|
| Best for | Keynotes, guest lectures, town halls | Welcomes, product films, stand loops | Receptions, stand hosts, product questions |
| Can it respond? | Yes, the person is there | No, it plays as filmed | Yes, from your approved material |
| What it needs | A studio, a green screen or controlled backdrop, and a 100 Mbps uplink | A filming session, then local playback | A trained avatar and your documents |
| Real example | Hitachi, IMT Ghaziabad | Monotech, Shahi Exports | Maruti Suzuki, Race Eco Chain |
Live: when the person matters
At Hitachi's HR Convention in New Delhi, an executive in Tokyo addressed the stage as a live, life-size hologram, with no flights. At IMT Ghaziabad, two guest speakers appeared live in a PGDM classroom, with two-way interaction. Choose live if the audience should be able to question that person.
Recorded: when the message is fixed
Monotech's CEO welcomed every IMTOS 2025 visitor for four days. He was shot once, and the film played on a loop with no operator. Shahi Exports showed its collection at Bharat Tex 2026 in the same way: life-size model walks and garments, on loop for four days. Choose recorded when you want the same message for everyone, and no one to run it.
AI: when the questions keep changing
At MQFC 2.0, an AI avatar took unscripted dealer questions live, in English and Japanese. At Maruti Suzuki's vendor conference, an on-premise avatar answered questions on 750+ DigiLib projects at Bharat Mandapam. Race Eco Chain's Maya greets registered visitors by name and tells their host. Choose AI when no script could cover what people will ask.
You can combine them
One HoloBox can do more than one. NatWest Group's event ran a touch menu, an AI avatar and a remote banker live, and TechHR India '26 ran AI avatars, a live beam-in and a multi-avatar concierge on one unit.
Questions, answered.
Do you make the content?
Yes. Our XR Design Centre in Noida makes content for the HoloBox. It's priced separately from the rental.
Which languages can the AI speak?
English, Hindi, Marathi, Bengali, Urdu, Punjabi, French, German and more.
Does an AI avatar need internet?
Not necessarily. AI can run on-premise and offline, with models trained on-premise.
How long does it take to make content for a hologram display?
Content made for you usually takes 2 to 4 weeks. A live telepresence session needs no production, only a camera at the sending end. A leadership message can be built from photos, and AI avatars answer from your own documents.
Last reviewed 5 November 2026.
Tell us who should appear.
We'll say which format fits, and what it needs.