Presence on Demand: Giving Personal AI Agents a Face Without a Body

AI agents can now do real work for us, such as sorting email, scheduling meetings and posting updates. Most of this work happens out of sight, so it is hard to know what the agent did, whether to trust it, or how to give it new instructions. This paper proposes “presence on demand”: a small pocket device keeps an AI agent working in the background, and a human-like AR avatar appears only when the user presses a button, reports what happened and takes new instructions. The paper describes the design, a layered safety approach (permission tiers, a link-opening rule and an independent verifier agent), and a proposed user study to test whether the avatar improves trust and error detection or only makes people over-trust the agent. This is a design paper with a proposed evaluation. The study has not yet been run.

Authors

Institutions

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-10-31
DOI
https://doi.org/10.5281/zenodo.23841643
Primary Topic
Social Robot Interaction and HRI
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
preprint

Presence on Demand: Giving Personal AI Agents a Face Without a Body

FAIZAN TOHEED
Zenodo (CERN European Organization for Nuclear Research)
Social Robot Interaction and HRI
preprint

Presence on Demand: Giving Personal AI Agents a Face Without a Body

FAIZAN TOHEED
preprint en

Abstract

AI agents can now do real work for us, such as sorting email, scheduling meetings and posting updates. Most of this work happens out of sight, so it is hard to know what the agent did, whether to trust it, or how to give it new instructions. This paper proposes “presence on demand”: a small pocket device keeps an AI agent working in the background, and a human-like AR avatar appears only when the user presses a button, reports what happened and takes new instructions. The paper describes the design, a layered safety approach (permission tiers, a link-opening rule and an independent verifier agent), and a proposed user study to test whether the avatar improves trust and error detection or only makes people over-trust the agent. This is a design paper with a proposed evaluation. The study has not yet been run.

Zenodo (CERN European Organization for Nuclear Research)
University of Management and Technology (PK)
Social Robot Interaction and HRI
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon